Add COPR badge

I'm not a RPM pro, so patches welcome! I'm surely doing something wrong.

.gitignore

@@ -1,3 +1,9 @@
+.omv/
+.ssh/
+.vagrant/
 mgmt-documentation.pdf
 old/
 tmp/
+*_stringer.go
+mgmt
+rpmbuild/

.travis.yml

@@ -0,0 +1,25 @@
+language: go
+go:
+  - 1.4.3
+  - 1.5.2
+  - tip
+dist: trusty
+sudo: required
+install: 'make deps'
+script: 'make test'
+matrix:
+  allow_failures:
+    - go: tip
+notifications:
+  irc:
+    channels:
+      - "irc.freenode.net#mgmtconfig"
+    template:
+      - "%{repository} (%{commit}: %{author}): %{message}"
+      - "More info : %{build_url}"
+    on_success: always
+    on_failure: always
+    use_notice: false
+    skip_join: false
+  email:
+    - travis-ci@shubin.ca

COPYRIGHT

@@ -1,5 +1,5 @@
 Mgmt
-Copyright (C) 2013-2015+ James Shubin and the project contributors
+Copyright (C) 2013-2016+ James Shubin and the project contributors
 Written by James Shubin <james@shubin.ca> and the project contributors
 
 This program is free software: you can redistribute it and/or modify

DOCUMENTATION.md

@@ -2,7 +2,7 @@
 
 <!--
 Mgmt
-Copyright (C) 2013-2015+ James Shubin and the project contributors
+Copyright (C) 2013-2016+ James Shubin and the project contributors
 Written by James Shubin <james@shubin.ca> and the project contributors
 
 This program is free software: you can redistribute it and/or modify
@@ -67,6 +67,14 @@ I wanted a next generation config management solution that didn't have all of
 the design flaws or limitations that the current generation of tools do, and no
 tool existed!
 
+###Why did you use etcd? What about consul?
+
+Etcd and consul are both written in golang, which made them the top two
+contenders for my prototype. Ultimately a choice had to be made, and etcd was
+chosen, but it was also somewhat arbitrary. If there is available interest,
+good reasoning, *and* patches, then we would consider either switching or
+supporting both, but this is not a high priority at this time.
+
 ###You didn't answer my question, or I have a question!
 
 It's best to ask on [IRC](https://webchat.freenode.net/?channels=#mgmtconfig)
@@ -99,6 +107,13 @@ documentation, please run `mgmt --help`.
 ####`--file <graph.yaml>`
 Point to a graph file to run.
 
+####`--converged-timeout <seconds>`
+Exit if the machine has converged for approximately this many seconds.
+
+####`--max-runtime <seconds>`
+Exit when the agent has run for approximately this many seconds. This is not
+generally recommended, but may be useful for users who know what they're doing.
+
 ##Examples
 For example configurations, please consult the [examples/](https://github.com/purpleidea/mgmt/tree/master/examples) directory in the git
 source repository. It is available from:
@@ -117,7 +132,7 @@ To report any bugs, please file a ticket at: [https://github.com/purpleidea/mgmt
 
 ##Authors
 
-Copyright (C) 2013-2015+ James Shubin and the project contributors
+Copyright (C) 2013-2016+ James Shubin and the project contributors
 
 Please see the
 [AUTHORS](https://github.com/purpleidea/mgmt/tree/master/AUTHORS) file

Makefile

@@ -1,9 +1,41 @@
+# Mgmt
+# Copyright (C) 2013-2016+ James Shubin and the project contributors
+# Written by James Shubin <james@shubin.ca> and the project contributors
+#
+# This program is free software: you can redistribute it and/or modify
+# it under the terms of the GNU Affero General Public License as published by
+# the Free Software Foundation, either version 3 of the License, or
+# (at your option) any later version.
+#
+# This program is distributed in the hope that it will be useful,
+# but WITHOUT ANY WARRANTY; without even the implied warranty of
+# MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
+# GNU Affero General Public License for more details.
+#
+# You should have received a copy of the GNU Affero General Public License
+# along with this program.  If not, see <http://www.gnu.org/licenses/>.
+
 SHELL = /bin/bash
-.PHONY: all version run race build clean test format docs
+.PHONY: all version program path deps run race build clean test format docs rpmbuild rpm srpm spec tar upload upload-sources upload-srpms upload-rpms copr
 .SILENT: clean
 
-VERSION := $(shell git describe --match '[0-9]*\.[0-9]*\.[0-9]*' --tags --dirty)
-PROGRAM := $(notdir $(CURDIR))
+SVERSION := $(shell git describe --match '[0-9]*\.[0-9]*\.[0-9]*' --tags --dirty --always)
+VERSION := $(shell git describe --match '[0-9]*\.[0-9]*\.[0-9]*' --tags --abbrev=0)
+PROGRAM := $(shell basename --suffix=-$(VERSION) $(notdir $(CURDIR)))
+ifeq ($(VERSION),$(SVERSION))
+	RELEASE = 1
+else
+	RELEASE = untagged
+endif
+ARCH = $(shell arch)
+SPEC = rpmbuild/SPECS/mgmt.spec
+SOURCE = rpmbuild/SOURCES/mgmt-$(VERSION).tar.bz2
+SRPM = rpmbuild/SRPMS/mgmt-$(VERSION)-$(RELEASE).src.rpm
+SRPM_BASE = mgmt-$(VERSION)-$(RELEASE).src.rpm
+RPM = rpmbuild/RPMS/mgmt-$(VERSION)-$(RELEASE).$(ARCH).rpm
+USERNAME := $(shell cat ~/.config/copr 2>/dev/null | grep username | awk -F '=' '{print $$2}' | tr -d ' ')
+SERVER = 'dl.fedoraproject.org'
+REMOTE_PATH = 'pub/alt/$(USERNAME)/mgmt'
 
 all: docs
 
@@ -11,35 +43,179 @@ all: docs
 version:
 	@echo $(VERSION)
 
-run:
-	find -maxdepth 1 -type f -name '*.go' -not -name '*_test.go' | xargs go run -ldflags "-X main.version $(VERSION) -X main.program $(PROGRAM)"
+program:
+	@echo $(PROGRAM)
 
-# include race test
+path:
+	./misc/make-path.sh
+
+deps:
+	./misc/make-deps.sh
+
+run:
+	find -maxdepth 1 -type f -name '*.go' -not -name '*_test.go' | xargs go run -ldflags "-X main.program=$(PROGRAM) -X main.version=$(SVERSION)"
+
+# include race flag
 race:
-	find -maxdepth 1 -type f -name '*.go' -not -name '*_test.go' | xargs go run -race -ldflags "-X main.version $(VERSION) -X main.program $(PROGRAM)"
+	find -maxdepth 1 -type f -name '*.go' -not -name '*_test.go' | xargs go run -race -ldflags "-X main.program=$(PROGRAM) -X main.version=$(SVERSION)"
 
 build: mgmt
 
 mgmt: main.go
-	go build -ldflags "-X main.version $(VERSION) -X main.program $(PROGRAM)"
+	@echo "Building: $(PROGRAM), version: $(SVERSION)."
+	go generate
+	# avoid equals sign in old golang versions eg in: -X foo=bar
+	if go version | grep -qE 'go1.3|go1.4'; then \
+		go build -ldflags "-X main.program $(PROGRAM) -X main.version $(SVERSION)" -o mgmt; \
+	else \
+		go build -ldflags "-X main.program=$(PROGRAM) -X main.version=$(SVERSION)" -o mgmt; \
+	fi
 
 clean:
 	[ ! -e mgmt ] || rm mgmt
+	rm -f *_stringer.go	# generated by `go generate`
 
 test:
 	./test.sh
-	./test/test-gofmt.sh
-	./test/test-yamlfmt.sh
-	go test
-	#go test ./pgraph
-	go test -race
-	#go test -race ./pgraph
 
 format:
 	find -type f -name '*.go' -not -path './old/*' -not -path './tmp/*' -exec gofmt -w {} \;
-	find -type f -name '*.yaml' -not -path './old/*' -not -path './tmp/*' -exec ruby -e "require 'yaml'; x=YAML.load_file('{}').to_yaml; File.open('{}', 'w').write x" \;
+	find -type f -name '*.yaml' -not -path './old/*' -not -path './tmp/*' -not -path './omv.yaml' -exec ruby -e "require 'yaml'; x=YAML.load_file('{}').to_yaml.each_line.map(&:rstrip).join(10.chr)+10.chr; File.open('{}', 'w').write x" \;
 
 docs: mgmt-documentation.pdf
 
 mgmt-documentation.pdf: DOCUMENTATION.md
 	pandoc DOCUMENTATION.md -o 'mgmt-documentation.pdf'
+
+#
+#	build aliases
+#
+# TODO: does making an rpm depend on making a .srpm first ?
+rpm: $(SRPM) $(RPM)
+	# do nothing
+
+srpm: $(SRPM)
+	# do nothing
+
+spec: $(SPEC)
+	# do nothing
+
+tar: $(SOURCE)
+	# do nothing
+
+rpmbuild/SOURCES/: tar
+rpmbuild/SRPMS/: srpm
+rpmbuild/RPMS/: rpm
+
+upload: upload-sources upload-srpms upload-rpms
+	# do nothing
+
+#
+#	rpmbuild
+#
+$(RPM): $(SPEC) $(SOURCE)
+	@echo Running rpmbuild -bb...
+	rpmbuild --define '_topdir $(shell pwd)/rpmbuild' -bb $(SPEC) && \
+	mv rpmbuild/RPMS/$(ARCH)/mgmt-$(VERSION)-$(RELEASE).*.rpm $(RPM)
+
+$(SRPM): $(SPEC) $(SOURCE)
+	@echo Running rpmbuild -bs...
+	rpmbuild --define '_topdir $(shell pwd)/rpmbuild' -bs $(SPEC)
+	# renaming is not needed because we aren't using the dist variable
+	#mv rpmbuild/SRPMS/mgmt-$(VERSION)-$(RELEASE).*.src.rpm $(SRPM)
+
+#
+#	spec
+#
+$(SPEC): rpmbuild/ mgmt.spec.in
+	@echo Running templater...
+	#cat mgmt.spec.in > $(SPEC)
+	sed -e s/__VERSION__/$(VERSION)/ -e s/__RELEASE__/$(RELEASE)/ < mgmt.spec.in > $(SPEC)
+	# append a changelog to the .spec file
+	git log --format="* %cd %aN <%aE>%n- (%h) %s%d%n" --date=local | sed -r 's/[0-9]+:[0-9]+:[0-9]+ //' >> $(SPEC)
+
+#
+#	archive
+#
+$(SOURCE): rpmbuild/
+	@echo Running git archive...
+	# use HEAD if tag doesn't exist yet, so that development is easier...
+	git archive --prefix=mgmt-$(VERSION)/ -o $(SOURCE) $(VERSION) 2> /dev/null || (echo 'Warning: $(VERSION) does not exist. Using HEAD instead.' && git archive --prefix=mgmt-$(VERSION)/ -o $(SOURCE) HEAD)
+	# TODO: if git archive had a --submodules flag this would easier!
+	@echo Running git archive submodules...
+	# i thought i would need --ignore-zeros, but it doesn't seem necessary!
+	p=`pwd` && (echo .; git submodule foreach) | while read entering path; do \
+		temp="$${path%\'}"; \
+		temp="$${temp#\'}"; \
+		path=$$temp; \
+		[ "$$path" = "" ] && continue; \
+		(cd $$path && git archive --prefix=mgmt-$(VERSION)/$$path/ HEAD > $$p/rpmbuild/tmp.tar && tar --concatenate --file=$$p/$(SOURCE) $$p/rpmbuild/tmp.tar && rm $$p/rpmbuild/tmp.tar); \
+	done
+
+# TODO: ensure that each sub directory exists
+rpmbuild/:
+	mkdir -p rpmbuild/{BUILD,BUILDROOT,RPMS,SOURCES,SPECS,SRPMS}
+
+rpmbuild:
+	mkdir -p rpmbuild/{BUILD,BUILDROOT,RPMS,SOURCES,SPECS,SRPMS}
+
+#
+#	sha256sum
+#
+rpmbuild/SOURCES/SHA256SUMS: rpmbuild/SOURCES/ $(SOURCE)
+	@echo Running SOURCES sha256sum...
+	cd rpmbuild/SOURCES/ && sha256sum *.tar.bz2 > SHA256SUMS; cd -
+
+rpmbuild/SRPMS/SHA256SUMS: rpmbuild/SRPMS/ $(SRPM)
+	@echo Running SRPMS sha256sum...
+	cd rpmbuild/SRPMS/ && sha256sum *src.rpm > SHA256SUMS; cd -
+
+rpmbuild/RPMS/SHA256SUMS: rpmbuild/RPMS/ $(RPM)
+	@echo Running RPMS sha256sum...
+	cd rpmbuild/RPMS/ && sha256sum *.rpm > SHA256SUMS; cd -
+
+#
+#	gpg
+#
+rpmbuild/SOURCES/SHA256SUMS.asc: rpmbuild/SOURCES/SHA256SUMS
+	@echo Running SOURCES gpg...
+	# the --yes forces an overwrite of the SHA256SUMS.asc if necessary
+	gpg2 --yes --clearsign rpmbuild/SOURCES/SHA256SUMS
+
+rpmbuild/SRPMS/SHA256SUMS.asc: rpmbuild/SRPMS/SHA256SUMS
+	@echo Running SRPMS gpg...
+	gpg2 --yes --clearsign rpmbuild/SRPMS/SHA256SUMS
+
+rpmbuild/RPMS/SHA256SUMS.asc: rpmbuild/RPMS/SHA256SUMS
+	@echo Running RPMS gpg...
+	gpg2 --yes --clearsign rpmbuild/RPMS/SHA256SUMS
+
+#
+#	upload
+#
+# upload to public server
+upload-sources: rpmbuild/SOURCES/ rpmbuild/SOURCES/SHA256SUMS rpmbuild/SOURCES/SHA256SUMS.asc
+	if [ "`cat rpmbuild/SOURCES/SHA256SUMS`" != "`ssh $(SERVER) 'cd $(REMOTE_PATH)/SOURCES/ && cat SHA256SUMS'`" ]; then \
+		echo Running SOURCES upload...; \
+		rsync -avz rpmbuild/SOURCES/ $(SERVER):$(REMOTE_PATH)/SOURCES/; \
+	fi
+
+upload-srpms: rpmbuild/SRPMS/ rpmbuild/SRPMS/SHA256SUMS rpmbuild/SRPMS/SHA256SUMS.asc
+	if [ "`cat rpmbuild/SRPMS/SHA256SUMS`" != "`ssh $(SERVER) 'cd $(REMOTE_PATH)/SRPMS/ && cat SHA256SUMS'`" ]; then \
+		echo Running SRPMS upload...; \
+		rsync -avz rpmbuild/SRPMS/ $(SERVER):$(REMOTE_PATH)/SRPMS/; \
+	fi
+
+upload-rpms: rpmbuild/RPMS/ rpmbuild/RPMS/SHA256SUMS rpmbuild/RPMS/SHA256SUMS.asc
+	if [ "`cat rpmbuild/RPMS/SHA256SUMS`" != "`ssh $(SERVER) 'cd $(REMOTE_PATH)/RPMS/ && cat SHA256SUMS'`" ]; then \
+		echo Running RPMS upload...; \
+		rsync -avz --prune-empty-dirs rpmbuild/RPMS/ $(SERVER):$(REMOTE_PATH)/RPMS/; \
+	fi
+
+#
+#	copr build
+#
+copr: upload-srpms
+	./misc/copr-build.py https://$(SERVER)/$(REMOTE_PATH)/SRPMS/$(SRPM_BASE)
+
+# vim: ts=8

README.md

@@ -1,35 +1,65 @@
 # *mgmt*: This is: mgmt!
 
-[![Build Status](https://secure.travis-ci.org/purpleidea/mgmt.png)](http://travis-ci.org/purpleidea/mgmt)
+[![Build Status](https://secure.travis-ci.org/purpleidea/mgmt.png?branch=master)](http://travis-ci.org/purpleidea/mgmt)
+[![Documentation](https://img.shields.io/docs/markdown.png)](DOCUMENTATION.md)
+[![IRC](https://img.shields.io/irc/%23mgmtconfig.png)](https://webchat.freenode.net/?channels=#mgmtconfig)
+[![Jenkins](https://img.shields.io/jenkins/status.png)](https://ci.centos.org/job/purpleidea-mgmt/)
+[![COPR](https://img.shields.io/copr/builds.png)](https://copr.fedoraproject.org/coprs/purpleidea/mgmt/)
+
+## Community:
+Come join us on IRC in [#mgmtconfig](https://webchat.freenode.net/?channels=#mgmtconfig) on Freenode!
+You may like the [#mgmtconfig](https://twitter.com/hashtag/mgmtconfig) hashtag if you're on [Twitter](https://twitter.com/#!/purpleidea).
+
+## Questions:
+Please join the [#mgmtconfig](https://webchat.freenode.net/?channels=#mgmtconfig) IRC community!
+If you have a well phrased question that might benefit others, consider asking it by sending a patch to the documentation [FAQ](https://github.com/purpleidea/mgmt/blob/master/DOCUMENTATION.md#usage-and-frequently-asked-questions) section. I'll merge your question, and a patch with the answer!
+
+## Quick start:
+* Either get the golang dependencies on your own, or run `make deps` if you're comfortable with how we install them.
+* Run `make build` to get a fresh built `mgmt` binary.
+* Run `cd $(mktemp --tmpdir -d tmp.XXX) && etcd` to get etcd running. The `mgmt` software will do this automatically for you in the future.
+* Run `time ./mgmt run --file examples/graph0.yaml --converged-timeout=1` to try out a very simple example!
+* To run continuously in the default mode of operation, omit the `--converged-timeout` option.
+* Have fun hacking on our future technology!
+
+## Examples:
+Please look in the [examples/](examples/) folder for more examples!
 
 ## Documentation:
 Please see: [DOCUMENTATION.md](DOCUMENTATION.md) or [PDF](https://pdfdoc-purpleidea.rhcloud.com/pdf/https://github.com/purpleidea/mgmt/blob/master/DOCUMENTATION.md).
 
-## Questions:
-Come join us in [#mgmtconfig](https://webchat.freenode.net/?channels=#mgmtconfig) on Freenode!
+## Roadmap:
+Please see: [TODO.md](TODO.md) for a list of upcoming work and TODO items.
+Please get involved by working on one of these items or by suggesting something else!
 
-## Examples:
-Please look in the [examples/](examples/) folder for usage. If none exist, please contribute one!
-
-## Notes:
-* This is currently a research project into next generation config management technologies!
-* This is my first complex project in golang, please notify me of any issues.
-* I have some well thought out designs for the future of this project, which I'll try and write up clearly and publish as soon as possible.
-* Please don't expect stable interfaces, code, or any data safety.
-* This design is the result of ideas I've had from hacking on advanced config management projects.
-* I first started hacking on this in ~2013, even though I had very little time for it.
-* I couldn't think of a good name for the project, so it's now being called `mgmt` until someone contributes a better one!
-* I've published a number of articles about this tool:
-  * TODO
-* There are some screencasts available:
-  * TODO
+## Bugs:
+Please set the `DEBUG` constant in [main.go](https://github.com/purpleidea/mgmt/blob/master/main.go) to `true`, and post the logs when you report the [issue](https://github.com/purpleidea/mgmt/issues).
+Bonus points if you provide a [shell](https://github.com/purpleidea/mgmt/tree/master/test/shell) or [OMV](https://github.com/purpleidea/mgmt/tree/master/test/omv) reproducible test case.
 
 ## Dependencies:
-* golang (available in most distros)
-* pandoc (for building a pdf of the documentation)
+* golang 1.4 or higher (required, available in most distros)
+* golang libraries (required, available with `go get`)
+
+        go get github.com/coreos/etcd/client
+        go get gopkg.in/yaml.v2
+        go get gopkg.in/fsnotify.v1
+        go get github.com/codegangsta/cli
+        go get github.com/coreos/go-systemd/dbus
+        go get github.com/coreos/go-systemd/util
+
+* stringer (required for building), available as a package on some platforms, otherwise via `go get`
+
+        go get golang.org/x/tools/cmd/stringer
+
+* pandoc (optional, for building a pdf of the documentation)
+* graphviz (optional, for building a visual representation of the graph)
 
 ## Patches:
-We'd love to have your patch! Please send it by email, or as a pull request.
+We'd love to have your patches! Please send them by email, or as a pull request.
+
+## On the web:
+* Introductory blog post: [https://ttboj.wordpress.com/2016/01/18/next-generation-configuration-mgmt/](https://ttboj.wordpress.com/2016/01/18/next-generation-configuration-mgmt/)
+* Julian Dunn at Cfgmgmtcamp 2016 [https://www.youtube.com/watch?v=kfF9IATUask&t=1949&html5=1](https://www.youtube.com/watch?v=kfF9IATUask&t=1949&html5=1)
 
 ##
 

THANKS

@@ -9,6 +9,8 @@ Chris Wright - For encouraging me to continue work on my prototype.
 
 Daniel Riek - For supporting and sheltering this project from bureaucracy.
 
+Ira Cooper - For having an algorithmic design discussion with me.
+
 Jeff Darcy - For some algorithm recommendations, and NACKing my TopoSort idea!
 
 Red Hat, inc. - For paying my salary, thus financially supporting my hacking.

TODO.md

@@ -0,0 +1,43 @@
+# TODO
+If you're looking for something to do, look here!
+Let us know if you're working on one of the items.
+
+## Package resource
+- [ ] base type [bug](https://github.com/purpleidea/mgmt/issues/11)
+- [ ] dnf blocker [bug](https://github.com/hughsie/PackageKit/issues/110)
+- [ ] install signal blocker [bug](https://github.com/hughsie/PackageKit/issues/109)
+
+## File resource
+- [ ] ability to make/delete folders
+- [ ] recursive argument (can recursively watch/modify contents)
+- [ ] force argument (can cause switch from file <-> folder)
+- [ ] fanotify support [bug](https://github.com/go-fsnotify/fsnotify/issues/114)
+
+## Exec resource
+- [ ] base resource improvements
+
+## Timer resource
+- [ ] base resource
+- [ ] reset on recompile
+- [ ] increment algorithm (linear, exponential, etc...)
+
+## Etcd improvements
+- [ ] embedded etcd master
+- [ ] capnslog fixes [bug](https://github.com/coreos/etcd/issues/4115)
+
+## Language improvements
+- [ ] language design
+- [ ] lexer/parser
+- [ ] automatic language formatter, ala `gofmt`
+- [ ] gedit/gnome-builder/gtksourceview syntax highlighting
+- [ ] vim syntax highlighting
+- [ ] emacs syntax highlighting
+
+## Other
+- [ ] better error/retry handling
+- [ ] resource grouping
+- [ ] automatic dependency adding (eg: packagekit file dependencies)
+- [ ] rpm package target in Makefile
+- [ ] deb package target in Makefile
+- [ ] reproducible builds
+- [ ] add your suggestions!

config.go

@@ -1,5 +1,5 @@
 // Mgmt
-// Copyright (C) 2013-2015+ James Shubin and the project contributors
+// Copyright (C) 2013-2016+ James Shubin and the project contributors
 // Written by James Shubin <james@shubin.ca> and the project contributors
 //
 // This program is free software: you can redistribute it and/or modify
@@ -22,23 +22,12 @@ import (
 	"gopkg.in/yaml.v2"
 	"io/ioutil"
 	"log"
+	"strings"
 )
 
-type noopTypeConfig struct {
-	Name string `yaml:"name"`
-}
-
-type fileTypeConfig struct {
-	Name    string `yaml:"name"`
-	Path    string `yaml:"path"`
-	Content string `yaml:"content"`
-	State   string `yaml:"state"`
-}
-
-type serviceTypeConfig struct {
-	Name    string `yaml:"name"`
-	State   string `yaml:"state"`
-	Startup string `yaml:"startup"`
+type collectorTypeConfig struct {
+	Type    string `yaml:"type"`
+	Pattern string `yaml:"pattern"` // XXX: Not Implemented
 }
 
 type vertexConfig struct {
@@ -52,18 +41,20 @@ type edgeConfig struct {
 	To   vertexConfig `yaml:"to"`
 }
 
-type graphConfig struct {
+type GraphConfig struct {
 	Graph string `yaml:"graph"`
 	Types struct {
-		Noop    []noopTypeConfig    `yaml:"noop"`
-		File    []fileTypeConfig    `yaml:"file"`
-		Service []serviceTypeConfig `yaml:"service"`
+		Noop    []NoopType    `yaml:"noop"`
+		File    []FileType    `yaml:"file"`
+		Service []ServiceType `yaml:"service"`
+		Exec    []ExecType    `yaml:"exec"`
 	} `yaml:"types"`
-	Edges   []edgeConfig `yaml:"edges"`
-	Comment string       `yaml:"comment"`
+	Collector []collectorTypeConfig `yaml:"collect"`
+	Edges     []edgeConfig          `yaml:"edges"`
+	Comment   string                `yaml:"comment"`
 }
 
-func (c *graphConfig) Parse(data []byte) error {
+func (c *GraphConfig) Parse(data []byte) error {
 	if err := yaml.Unmarshal(data, c); err != nil {
 		return err
 	}
@@ -73,54 +64,166 @@ func (c *graphConfig) Parse(data []byte) error {
 	return nil
 }
 
-func GraphFromConfig(filename string) *Graph {
+func ParseConfigFromFile(filename string) *GraphConfig {
+	data, err := ioutil.ReadFile(filename)
+	if err != nil {
+		log.Printf("Error: Config: ParseConfigFromFile: File: %v", err)
+		return nil
+	}
 
-	var NoopMap map[string]*Vertex = make(map[string]*Vertex)
-	var FileMap map[string]*Vertex = make(map[string]*Vertex)
-	var ServiceMap map[string]*Vertex = make(map[string]*Vertex)
+	var config GraphConfig
+	if err := config.Parse(data); err != nil {
+		log.Printf("Error: Config: ParseConfigFromFile: Parse: %v", err)
+		return nil
+	}
 
-	var lookup map[string]map[string]*Vertex = make(map[string]map[string]*Vertex)
+	return &config
+}
+
+// XXX: we need to fix this function so that it either fails without modifying
+// the graph, passes successfully and modifies it, or basically panics i guess
+// this way an invalid compilation can leave the old graph running, and we we
+// don't modify a partial graph. so we really need to validate, and then perform
+// whatever actions are necessary
+// finding some way to do this on a copy of the graph, and then do a graph diff
+// and merge the new data into the old graph would be more appropriate, in
+// particular if we can ensure the graph merge can't fail. As for the putting
+// of stuff into etcd, we should probably store the operations to complete in
+// the new graph, and keep retrying until it succeeds, thus blocking any new
+// etcd operations until that time.
+func UpdateGraphFromConfig(config *GraphConfig, hostname string, g *Graph, etcdO *EtcdWObject) bool {
+
+	var NoopMap = make(map[string]*Vertex)
+	var FileMap = make(map[string]*Vertex)
+	var ServiceMap = make(map[string]*Vertex)
+	var ExecMap = make(map[string]*Vertex)
+
+	var lookup = make(map[string]map[string]*Vertex)
 	lookup["noop"] = NoopMap
 	lookup["file"] = FileMap
 	lookup["service"] = ServiceMap
+	lookup["exec"] = ExecMap
 
-	data, err := ioutil.ReadFile(filename)
-	if err != nil {
-		log.Fatal(err)
-	}
+	//log.Printf("%+v", config) // debug
 
-	var config graphConfig
-	if err := config.Parse(data); err != nil {
-		log.Fatal(err)
-	}
-	//fmt.Printf("%+v\n", config)	// debug
+	g.SetName(config.Graph) // set graph name
 
-	g := NewGraph(config.Graph)
+	var keep []*Vertex // list of vertex which are the same in new graph
 
 	for _, t := range config.Types.Noop {
-		NoopMap[t.Name] = NewVertex(t.Name, "noop")
-		// FIXME: duplicate of name stored twice... where should it go?
-		NoopMap[t.Name].Associate(NewNoopType(t.Name))
-		g.AddVertex(NoopMap[t.Name]) // call standalone in case not part of an edge
+		obj := NewNoopType(t.Name)
+		v := g.GetVertexMatch(obj)
+		if v == nil { // no match found
+			v = NewVertex(obj)
+			g.AddVertex(v) // call standalone in case not part of an edge
+		}
+		NoopMap[obj.Name] = v  // used for constructing edges
+		keep = append(keep, v) // append
 	}
 
 	for _, t := range config.Types.File {
-		FileMap[t.Name] = NewVertex(t.Name, "file")
-		// FIXME: duplicate of name stored twice... where should it go?
-		FileMap[t.Name].Associate(NewFileType(t.Name, t.Path, t.Content, t.State))
-		g.AddVertex(FileMap[t.Name]) // call standalone in case not part of an edge
+		// XXX: should we export based on a @@ prefix, or a metaparam
+		// like exported => true || exported => (host pattern)||(other pattern?)
+		if strings.HasPrefix(t.Name, "@@") { // exported resource
+			// add to etcd storage...
+			t.Name = t.Name[2:] //slice off @@
+			if !etcdO.EtcdPut(hostname, t.Name, "file", t) {
+				log.Printf("Problem exporting file resource %v.", t.Name)
+				continue
+			}
+		} else {
+			obj := NewFileType(t.Name, t.Path, t.Dirname, t.Basename, t.Content, t.State)
+			v := g.GetVertexMatch(obj)
+			if v == nil { // no match found
+				v = NewVertex(obj)
+				g.AddVertex(v) // call standalone in case not part of an edge
+			}
+			FileMap[obj.Name] = v  // used for constructing edges
+			keep = append(keep, v) // append
+		}
 	}
 
 	for _, t := range config.Types.Service {
-		ServiceMap[t.Name] = NewVertex(t.Name, "service")
-		// FIXME: duplicate of name stored twice... where should it go?
-		ServiceMap[t.Name].Associate(NewServiceType(t.Name, t.State, t.Startup))
-		g.AddVertex(ServiceMap[t.Name]) // call standalone in case not part of an edge
+		obj := NewServiceType(t.Name, t.State, t.Startup)
+		v := g.GetVertexMatch(obj)
+		if v == nil { // no match found
+			v = NewVertex(obj)
+			g.AddVertex(v) // call standalone in case not part of an edge
+		}
+		ServiceMap[obj.Name] = v // used for constructing edges
+		keep = append(keep, v)   // append
+	}
+
+	for _, t := range config.Types.Exec {
+		obj := NewExecType(t.Name, t.Cmd, t.Shell, t.Timeout, t.WatchCmd, t.WatchShell, t.IfCmd, t.IfShell, t.PollInt, t.State)
+		v := g.GetVertexMatch(obj)
+		if v == nil { // no match found
+			v = NewVertex(obj)
+			g.AddVertex(v) // call standalone in case not part of an edge
+		}
+		ExecMap[obj.Name] = v  // used for constructing edges
+		keep = append(keep, v) // append
+	}
+
+	// lookup from etcd graph
+	// do all the graph look ups in one single step, so that if the etcd
+	// database changes, we don't have a partial state of affairs...
+	nodes, ok := etcdO.EtcdGet()
+	if ok {
+		for _, t := range config.Collector {
+			// XXX: use t.Type and optionally t.Pattern to collect from etcd storage
+			log.Printf("Collect: %v; Pattern: %v", t.Type, t.Pattern)
+
+			for _, x := range etcdO.EtcdGetProcess(nodes, "file") {
+				var obj *FileType
+				if B64ToObj(x, &obj) != true {
+					log.Printf("Collect: File: %v not collected!", x)
+					continue
+				}
+				if t.Pattern != "" { // XXX: currently the pattern for files can only override the Dirname variable :P
+					obj.Dirname = t.Pattern
+				}
+
+				log.Printf("Collect: File: %v collected!", obj.GetName())
+
+				// XXX: similar to file add code:
+				v := g.GetVertexMatch(obj)
+				if v == nil { // no match found
+					obj.Init() // initialize go channels or things won't work!!!
+					v = NewVertex(obj)
+					g.AddVertex(v) // call standalone in case not part of an edge
+				}
+				FileMap[obj.GetName()] = v // used for constructing edges
+				keep = append(keep, v)     // append
+
+			}
+
+		}
+	}
+
+	// get rid of any vertices we shouldn't "keep" (that aren't in new graph)
+	for _, v := range g.GetVertices() {
+		if !HasVertex(v, keep) {
+			// wait for exit before starting new graph!
+			v.Type.SendEvent(eventExit, true, false)
+			g.DeleteVertex(v)
+		}
 	}
 
 	for _, e := range config.Edges {
+		if _, ok := lookup[e.From.Type]; !ok {
+			return false
+		}
+		if _, ok := lookup[e.To.Type]; !ok {
+			return false
+		}
+		if _, ok := lookup[e.From.Type][e.From.Name]; !ok {
+			return false
+		}
+		if _, ok := lookup[e.To.Type][e.To.Name]; !ok {
+			return false
+		}
 		g.AddEdge(lookup[e.From.Type][e.From.Name], lookup[e.To.Type][e.To.Name], NewEdge(e.Name))
 	}
-
-	return g
+	return true
 }

configwatch.go

@@ -0,0 +1,155 @@
+// Mgmt
+// Copyright (C) 2013-2016+ James Shubin and the project contributors
+// Written by James Shubin <james@shubin.ca> and the project contributors
+//
+// This program is free software: you can redistribute it and/or modify
+// it under the terms of the GNU Affero General Public License as published by
+// the Free Software Foundation, either version 3 of the License, or
+// (at your option) any later version.
+//
+// This program is distributed in the hope that it will be useful,
+// but WITHOUT ANY WARRANTY; without even the implied warranty of
+// MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
+// GNU Affero General Public License for more details.
+//
+// You should have received a copy of the GNU Affero General Public License
+// along with this program.  If not, see <http://www.gnu.org/licenses/>.
+
+package main
+
+import (
+	"gopkg.in/fsnotify.v1"
+	//"github.com/go-fsnotify/fsnotify" // git master of "gopkg.in/fsnotify.v1"
+	"log"
+	"math"
+	"path"
+	"strings"
+	"syscall"
+)
+
+// XXX: it would be great if we could reuse code between this and the file type
+// XXX: patch this to submit it as part of go-fsnotify if they're interested...
+func ConfigWatch(file string) chan bool {
+	ch := make(chan bool)
+	go func() {
+		var safename = path.Clean(file) // no trailing slash
+
+		watcher, err := fsnotify.NewWatcher()
+		if err != nil {
+			log.Fatal(err)
+		}
+		defer watcher.Close()
+
+		patharray := PathSplit(safename) // tokenize the path
+		var index = len(patharray)       // starting index
+		var current string               // current "watcher" location
+		var deltaDepth int               // depth delta between watcher and event
+		var send = false                 // send event?
+
+		for {
+			current = strings.Join(patharray[0:index], "/")
+			if current == "" { // the empty string top is the root dir ("/")
+				current = "/"
+			}
+			log.Printf("Watching: %v", current) // attempting to watch...
+
+			// initialize in the loop so that we can reset on rm-ed handles
+			err = watcher.Add(current)
+			if err != nil {
+				if err == syscall.ENOENT {
+					index-- // usually not found, move up one dir
+				} else if err == syscall.ENOSPC {
+					// XXX: occasionally: no space left on device,
+					// XXX: probably due to lack of inotify watches
+					log.Printf("Lack of watches for config(%v) error: %+v", file, err.Error) // 0x408da0
+					log.Fatal(err)
+				} else {
+					log.Printf("Unknown config(%v) error:", file)
+					log.Fatal(err)
+				}
+				index = int(math.Max(1, float64(index)))
+				continue
+			}
+
+			select {
+			case event := <-watcher.Events:
+				// the deeper you go, the bigger the deltaDepth is...
+				// this is the difference between what we're watching,
+				// and the event... doesn't mean we can't watch deeper
+				if current == event.Name {
+					deltaDepth = 0 // i was watching what i was looking for
+
+				} else if HasPathPrefix(event.Name, current) {
+					deltaDepth = len(PathSplit(current)) - len(PathSplit(event.Name)) // -1 or less
+
+				} else if HasPathPrefix(current, event.Name) {
+					deltaDepth = len(PathSplit(event.Name)) - len(PathSplit(current)) // +1 or more
+
+				} else {
+					// TODO different watchers get each others events!
+					// https://github.com/go-fsnotify/fsnotify/issues/95
+					// this happened with two values such as:
+					// event.Name: /tmp/mgmt/f3 and current: /tmp/mgmt/f2
+					continue
+				}
+				//log.Printf("The delta depth is: %v", deltaDepth)
+
+				// if we have what we wanted, awesome, send an event...
+				if event.Name == safename {
+					//log.Println("Event!")
+					send = true
+
+					// file removed, move the watch upwards
+					if deltaDepth >= 0 && (event.Op&fsnotify.Remove == fsnotify.Remove) {
+						//log.Println("Removal!")
+						watcher.Remove(current)
+						index--
+					}
+
+					// we must be a parent watcher, so descend in
+					if deltaDepth < 0 {
+						watcher.Remove(current)
+						index++
+					}
+
+					// if safename starts with event.Name, we're above, and no event should be sent
+				} else if HasPathPrefix(safename, event.Name) {
+					//log.Println("Above!")
+
+					if deltaDepth >= 0 && (event.Op&fsnotify.Remove == fsnotify.Remove) {
+						log.Println("Removal!")
+						watcher.Remove(current)
+						index--
+					}
+
+					if deltaDepth < 0 {
+						log.Println("Parent!")
+						if PathPrefixDelta(safename, event.Name) == 1 { // we're the parent dir
+							//send = true
+						}
+						watcher.Remove(current)
+						index++
+					}
+
+					// if event.Name startswith safename, send event, we're already deeper
+				} else if HasPathPrefix(event.Name, safename) {
+					//log.Println("Event2!")
+					//send = true
+				}
+
+			case err := <-watcher.Errors:
+				log.Println("error:", err)
+				log.Fatal(err)
+
+			}
+
+			// do our event sending all together to avoid duplicate msgs
+			if send {
+				send = false
+				ch <- true
+			}
+		}
+		//close(ch)
+	}()
+	return ch
+}

etcd.go

@@ -0,0 +1,300 @@
+// Mgmt
+// Copyright (C) 2013-2016+ James Shubin and the project contributors
+// Written by James Shubin <james@shubin.ca> and the project contributors
+//
+// This program is free software: you can redistribute it and/or modify
+// it under the terms of the GNU Affero General Public License as published by
+// the Free Software Foundation, either version 3 of the License, or
+// (at your option) any later version.
+//
+// This program is distributed in the hope that it will be useful,
+// but WITHOUT ANY WARRANTY; without even the implied warranty of
+// MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
+// GNU Affero General Public License for more details.
+//
+// You should have received a copy of the GNU Affero General Public License
+// along with this program.  If not, see <http://www.gnu.org/licenses/>.
+
+package main
+
+import (
+	"fmt"
+	etcd_context "github.com/coreos/etcd/Godeps/_workspace/src/golang.org/x/net/context"
+	etcd "github.com/coreos/etcd/client"
+	"log"
+	"math"
+	"strings"
+	"time"
+)
+
+//go:generate stringer -type=etcdMsg -output=etcdmsg_stringer.go
+type etcdMsg int
+
+const (
+	etcdStart etcdMsg = iota
+	etcdEvent
+	etcdFoo
+	etcdBar
+)
+
+//go:generate stringer -type=etcdConvergedState -output=etcdconvergedstate_stringer.go
+type etcdConvergedState int
+
+const (
+	etcdConvergedNil etcdConvergedState = iota
+	//etcdConverged
+	etcdConvergedTimeout
+)
+
+type EtcdWObject struct { // etcd wrapper object
+	seed           string
+	ctimeout       int
+	converged      chan bool
+	kapi           etcd.KeysAPI
+	convergedState etcdConvergedState
+}
+
+func (etcdO *EtcdWObject) GetConvergedState() etcdConvergedState {
+	return etcdO.convergedState
+}
+
+func (etcdO *EtcdWObject) SetConvergedState(state etcdConvergedState) {
+	etcdO.convergedState = state
+}
+
+func (etcdO *EtcdWObject) GetKAPI() etcd.KeysAPI {
+	if etcdO.kapi != nil { // memoize
+		return etcdO.kapi
+	}
+
+	cfg := etcd.Config{
+		Endpoints: []string{etcdO.seed},
+		Transport: etcd.DefaultTransport,
+		// set timeout per request to fail fast when the target endpoint is unavailable
+		HeaderTimeoutPerRequest: time.Second,
+	}
+
+	var c etcd.Client
+	var err error
+
+	c, err = etcd.New(cfg)
+	if err != nil {
+		// XXX: not sure if this ever errors
+		if cerr, ok := err.(*etcd.ClusterError); ok {
+			// XXX: not sure if this part ever matches
+			// not running or disconnected
+			if cerr == etcd.ErrClusterUnavailable {
+				log.Fatal("XXX: etcd: ErrClusterUnavailable")
+			} else {
+				log.Fatal("XXX: etcd: Unknown")
+			}
+		}
+		log.Fatal(err) // some unhandled error
+	}
+	etcdO.kapi = etcd.NewKeysAPI(c)
+	return etcdO.kapi
+}
+
+type EtcdChannelWatchResponse struct {
+	resp *etcd.Response
+	err  error
+}
+
+// wrap the etcd watcher.Next blocking function inside of a channel
+func (etcdO *EtcdWObject) EtcdChannelWatch(watcher etcd.Watcher, context etcd_context.Context) chan *EtcdChannelWatchResponse {
+	ch := make(chan *EtcdChannelWatchResponse)
+	go func() {
+		for {
+			resp, err := watcher.Next(context) // blocks here
+			ch <- &EtcdChannelWatchResponse{resp, err}
+		}
+	}()
+	return ch
+}
+
+func (etcdO *EtcdWObject) EtcdWatch() chan etcdMsg {
+	kapi := etcdO.GetKAPI()
+	ctimeout := etcdO.ctimeout
+	converged := etcdO.converged
+	// XXX: i think we need this buffered so that when we're hanging on the
+	// channel, which is inside the EtcdWatch main loop, we still want the
+	// calls to Get/Set on etcd to succeed, so blocking them here would
+	// kill the whole thing
+	ch := make(chan etcdMsg, 1) // XXX: buffer of at least 1 is required
+	go func(ch chan etcdMsg) {
+		tmin := 500   // initial (min) delay in ms
+		t := tmin     // current time
+		tmult := 2    // multiplier for exponential delay
+		tmax := 16000 // max delay
+		watcher := kapi.Watcher("/exported/", &etcd.WatcherOptions{Recursive: true})
+		etcdch := etcdO.EtcdChannelWatch(watcher, etcd_context.Background())
+		for {
+			log.Printf("Etcd: Watching...")
+			var resp *etcd.Response // = nil by default
+			var err error
+			select {
+			case out := <-etcdch:
+				etcdO.SetConvergedState(etcdConvergedNil)
+				resp, err = out.resp, out.err
+
+			case _ = <-TimeAfterOrBlock(ctimeout):
+				etcdO.SetConvergedState(etcdConvergedTimeout)
+				converged <- true
+				continue
+			}
+
+			if err != nil {
+				if err == etcd_context.Canceled {
+					// ctx is canceled by another routine
+					log.Fatal("Canceled")
+				} else if err == etcd_context.DeadlineExceeded {
+					// ctx is attached with a deadline and it exceeded
+					log.Fatal("Deadline")
+				} else if cerr, ok := err.(*etcd.ClusterError); ok {
+					// not running or disconnected
+					// TODO: is there a better way to parse errors?
+					for _, e := range cerr.Errors {
+						if strings.HasSuffix(e.Error(), "getsockopt: connection refused") {
+							t = int(math.Min(float64(t*tmult), float64(tmax)))
+							log.Printf("Etcd: Waiting %d ms for connection...", t)
+							time.Sleep(time.Duration(t) * time.Millisecond) // sleep for t ms
+
+						} else if e.Error() == "unexpected EOF" {
+							log.Printf("Etcd: Unexpected disconnect...")
+
+						} else if e.Error() == "EOF" {
+							log.Printf("Etcd: Disconnected...")
+
+						} else if strings.HasPrefix(e.Error(), "unsupported protocol scheme") {
+							// usually a bad peer endpoint value
+							log.Fatal("Bad peer endpoint value?")
+
+						} else {
+							log.Fatal("Woops: ", e.Error())
+						}
+					}
+				} else {
+					// bad cluster endpoints, which are not etcd servers
+					log.Fatal("Woops: ", err)
+				}
+			} else {
+				//log.Print(resp)
+				//log.Printf("Watcher().Node.Value(%v): %+v", key, resp.Node.Value)
+				// FIXME: we should actually reset when the server comes back, not here on msg!
+				//XXX: can we fix this with one of these patterns?: https://blog.golang.org/go-concurrency-patterns-timing-out-and
+				t = tmin // reset timer
+
+				// don't trigger event if nothing changed
+				if n, p := resp.Node, resp.PrevNode; resp.Action == "set" && p != nil {
+					if n.Key == p.Key && n.Value == p.Value {
+						continue
+					}
+				}
+
+				// FIXME: we get events on key/type/value changes for
+				// each type directory... ignore the non final ones...
+				// IOW, ignore everything except for the value or some
+				// field which gets set last... this could be the max count field thing...
+
+				log.Printf("Etcd: Value: %v", resp.Node.Value) // event
+				ch <- etcdEvent                                // event
+			}
+
+		} // end for loop
+		//close(ch)
+	}(ch) // call go routine
+	return ch
+}
+
+// helper function to store our data in etcd
+func (etcdO *EtcdWObject) EtcdPut(hostname, key, typ string, obj interface{}) bool {
+	kapi := etcdO.GetKAPI()
+	output, ok := ObjToB64(obj)
+	if !ok {
+		log.Printf("Etcd: Could not encode %v key.", key)
+		return false
+	}
+
+	path := fmt.Sprintf("/exported/%s/types/%s/type", hostname, key)
+	_, err := kapi.Set(etcd_context.Background(), path, typ, nil)
+	// XXX validate...
+
+	path = fmt.Sprintf("/exported/%s/types/%s/value", hostname, key)
+	resp, err := kapi.Set(etcd_context.Background(), path, output, nil)
+	if err != nil {
+		if cerr, ok := err.(*etcd.ClusterError); ok {
+			// not running or disconnected
+			for _, e := range cerr.Errors {
+				if strings.HasSuffix(e.Error(), "getsockopt: connection refused") {
+				}
+				//if e == etcd.ErrClusterUnavailable
+			}
+		}
+		log.Printf("Etcd: Could not store %v key.", key)
+		return false
+	}
+	log.Print("Etcd: ", resp) // w00t... bonus
+	return true
+}
+
+// lookup /exported/ node hierarchy
+func (etcdO *EtcdWObject) EtcdGet() (etcd.Nodes, bool) {
+	kapi := etcdO.GetKAPI()
+	// key structure is /exported/<hostname>/types/...
+	resp, err := kapi.Get(etcd_context.Background(), "/exported/", &etcd.GetOptions{Recursive: true})
+	if err != nil {
+		return nil, false // not found
+	}
+	return resp.Node.Nodes, true
+}
+
+func (etcdO *EtcdWObject) EtcdGetProcess(nodes etcd.Nodes, typ string) []string {
+	//path := fmt.Sprintf("/exported/%s/types/", h)
+	top := "/exported/"
+	log.Printf("Etcd: Get: %+v", nodes) // Get().Nodes.Nodes
+	var output []string
+
+	for _, x := range nodes { // loop through hosts
+		if !strings.HasPrefix(x.Key, top) {
+			log.Fatal("Error!")
+		}
+		host := x.Key[len(top):]
+		//log.Printf("Get().Nodes[%v]: %+v ==> %+v", -1, host, x.Nodes)
+		//log.Printf("Get().Nodes[%v]: %+v ==> %+v", i, x.Key, x.Nodes)
+		types, ok := EtcdGetChildNodeByKey(x, "types")
+		if !ok {
+			continue
+		}
+		for _, y := range types.Nodes { // loop through types
+			//key := y.Key # UUID?
+			//log.Printf("Get(%v): TYPE[%v]", host, y.Key)
+			t, ok := EtcdGetChildNodeByKey(y, "type")
+			if !ok {
+				continue
+			}
+			if typ != "" && typ != t.Value {
+				continue
+			} // filter based on type
+
+			v, ok := EtcdGetChildNodeByKey(y, "value") // B64ToObj this
+			if !ok {
+				continue
+			}
+			log.Printf("Etcd: Hostname: %v; Get: %v", host, t.Value)
+
+			output = append(output, v.Value)
+		}
+	}
+	return output
+}
+
+// TODO: wrap this somehow so it's a method of *etcd.Node
+// helper function that returns the node for a particular key under a node
+func EtcdGetChildNodeByKey(node *etcd.Node, key string) (*etcd.Node, bool) {
+	for _, x := range node.Nodes {
+		if x.Key == fmt.Sprintf("%s/%s", node.Key, key) {
+			return x, true
+		}
+	}
+	return nil, false // not found
+}

event.go

@@ -1,5 +1,5 @@
 // Mgmt
-// Copyright (C) 2013-2015+ James Shubin and the project contributors
+// Copyright (C) 2013-2016+ James Shubin and the project contributors
 // Written by James Shubin <james@shubin.ca> and the project contributors
 //
 // This program is free software: you can redistribute it and/or modify
@@ -17,20 +17,39 @@
 
 package main
 
-import (
-	"code.google.com/p/go-uuid/uuid"
+//go:generate stringer -type=eventName -output=eventname_stringer.go
+type eventName int
+
+const (
+	eventExit eventName = iota
+	eventStart
+	eventPause
+	eventPoke
+	eventBackPoke
 )
 
 type Event struct {
-	uuid string
-	Name string
-	Type string
+	Name eventName
+	Resp chan bool // channel to send an ack response on, nil to skip
+	//Wg   *sync.WaitGroup // receiver barrier to Wait() for everyone else on
+	Msg      string // some words for fun
+	Activity bool   // did something interesting happen?
 }
 
-func NewEvent(name, t string) *Event {
-	return &Event{
-		uuid: uuid.New(),
-		Name: name,
-		Type: t,
+// send a single acknowledgement on the channel if one was requested
+func (event *Event) ACK() {
+	if event.Resp != nil { // if they've requested an ACK
+		event.Resp <- true // send ACK
 	}
 }
+
+func (event *Event) NACK() {
+	if event.Resp != nil { // if they've requested an ACK
+		event.Resp <- false // send NACK
+	}
+}
+
+// get the activity value
+func (event *Event) GetActivity() bool {
+	return event.Activity
+}

examples/graph0.yaml

@@ -0,0 +1,20 @@
+---
+graph: mygraph
+comment: hello world example
+types:
+  noop:
+  - name: noop1
+  file:
+  - name: file1
+    path: "/tmp/mgmt-hello-world"
+    content: |
+      hello world from @purpleidea
+    state: exists
+edges:
+- name: e1
+  from:
+    type: noop
+    name: noop1
+  to:
+    type: file
+    name: file1

examples/graph1.yaml

@@ -5,22 +5,22 @@ types:
   - name: noop1
   file:
   - name: file1
-    path: /tmp/mgmt/f1
+    path: "/tmp/mgmt/f1"
     content: |
       i am f1
     state: exists
   - name: file2
-    path: /tmp/mgmt/f2
+    path: "/tmp/mgmt/f2"
     content: |
       i am f2
     state: exists
   - name: file3
-    path: /tmp/mgmt/f3
+    path: "/tmp/mgmt/f3"
     content: |
       i am f3
     state: exists
   - name: file4
-    path: /tmp/mgmt/f4
+    path: "/tmp/mgmt/f4"
     content: |
       i am f4 and i should not be here
     state: absent

examples/graph10.yaml

@@ -0,0 +1,128 @@
+---
+graph: mygraph
+comment: simple exec fan in to fan out example to demonstrate optimization
+types:
+  exec:
+  - name: exec1
+    cmd: sleep 10s
+    shell: ''
+    timeout: 0
+    watchcmd: ''
+    watchshell: ''
+    ifcmd: ''
+    ifshell: ''
+    pollint: 0
+    state: present
+  - name: exec2
+    cmd: sleep 10s
+    shell: ''
+    timeout: 0
+    watchcmd: ''
+    watchshell: ''
+    ifcmd: ''
+    ifshell: ''
+    pollint: 0
+    state: present
+  - name: exec3
+    cmd: sleep 10s
+    shell: ''
+    timeout: 0
+    watchcmd: ''
+    watchshell: ''
+    ifcmd: ''
+    ifshell: ''
+    pollint: 0
+    state: present
+  - name: exec4
+    cmd: sleep 10s
+    shell: ''
+    timeout: 0
+    watchcmd: ''
+    watchshell: ''
+    ifcmd: ''
+    ifshell: ''
+    pollint: 0
+    state: present
+  - name: exec5
+    cmd: sleep 10s
+    shell: ''
+    timeout: 0
+    watchcmd: ''
+    watchshell: ''
+    ifcmd: ''
+    ifshell: ''
+    pollint: 0
+    state: present
+  - name: exec6
+    cmd: sleep 10s
+    shell: ''
+    timeout: 0
+    watchcmd: ''
+    watchshell: ''
+    ifcmd: ''
+    ifshell: ''
+    pollint: 0
+    state: present
+  - name: exec7
+    cmd: sleep 10s
+    shell: ''
+    timeout: 0
+    watchcmd: ''
+    watchshell: ''
+    ifcmd: ''
+    ifshell: ''
+    pollint: 0
+    state: present
+  - name: exec8
+    cmd: sleep 15s
+    shell: ''
+    timeout: 0
+    watchcmd: ''
+    watchshell: ''
+    ifcmd: ''
+    ifshell: ''
+    pollint: 0
+    state: present
+edges:
+- name: e1
+  from:
+    type: exec
+    name: exec1
+  to:
+    type: exec
+    name: exec4
+- name: e2
+  from:
+    type: exec
+    name: exec2
+  to:
+    type: exec
+    name: exec4
+- name: e3
+  from:
+    type: exec
+    name: exec3
+  to:
+    type: exec
+    name: exec4
+- name: e4
+  from:
+    type: exec
+    name: exec4
+  to:
+    type: exec
+    name: exec5
+- name: e5
+  from:
+    type: exec
+    name: exec4
+  to:
+    type: exec
+    name: exec6
+- name: e6
+  from:
+    type: exec
+    name: exec4
+  to:
+    type: exec
+    name: exec7

examples/graph1a.yaml

@@ -0,0 +1,22 @@
+---
+graph: mygraph
+types:
+  file:
+  - name: file1
+    path: "/tmp/mgmt/f1"
+    content: |
+      i am f1
+    state: exists
+  - name: file2
+    path: "/tmp/mgmt/f2"
+    content: |
+      i am f2
+    state: exists
+edges:
+- name: e1
+  from:
+    type: file
+    name: file1
+  to:
+    type: file
+    name: file2

examples/graph1b.yaml

@@ -0,0 +1,22 @@
+---
+graph: mygraph
+types:
+  file:
+  - name: file2
+    path: "/tmp/mgmt/f2"
+    content: |
+      i am f2
+    state: exists
+  - name: file3
+    path: "/tmp/mgmt/f3"
+    content: |
+      i am f3
+    state: exists
+edges:
+- name: e2
+  from:
+    type: file
+    name: file2
+  to:
+    type: file
+    name: file3

examples/graph2.yaml

@@ -5,7 +5,7 @@ types:
   - name: noop1
   file:
   - name: file1
-    path: /tmp/mgmt/f1
+    path: "/tmp/mgmt/f1"
     content: |
       i am f1
     state: exists

examples/graph3a.yaml

@@ -0,0 +1,28 @@
+---
+graph: mygraph
+types:
+  file:
+  - name: file1a
+    path: "/tmp/mgmtA/f1a"
+    content: |
+      i am f1
+    state: exists
+  - name: file2a
+    path: "/tmp/mgmtA/f2a"
+    content: |
+      i am f2
+    state: exists
+  - name: "@@file3a"
+    path: "/tmp/mgmtA/f3a"
+    content: |
+      i am f3, exported from host A
+    state: exists
+  - name: "@@file4a"
+    path: "/tmp/mgmtA/f4a"
+    content: |
+      i am f4, exported from host A
+    state: exists
+collect:
+- type: file
+  pattern: "/tmp/mgmtA/"
+edges: []

examples/graph3b.yaml

@@ -0,0 +1,28 @@
+---
+graph: mygraph
+types:
+  file:
+  - name: file1b
+    path: "/tmp/mgmtB/f1b"
+    content: |
+      i am f1
+    state: exists
+  - name: file2b
+    path: "/tmp/mgmtB/f2b"
+    content: |
+      i am f2
+    state: exists
+  - name: "@@file3b"
+    path: "/tmp/mgmtB/f3b"
+    content: |
+      i am f3, exported from host B
+    state: exists
+  - name: "@@file4b"
+    path: "/tmp/mgmtB/f4b"
+    content: |
+      i am f4, exported from host B
+    state: exists
+collect:
+- type: file
+  pattern: "/tmp/mgmtB/"
+edges: []

examples/graph3c.yaml

@@ -0,0 +1,28 @@
+---
+graph: mygraph
+types:
+  file:
+  - name: file1c
+    path: "/tmp/mgmtC/f1c"
+    content: |
+      i am f1
+    state: exists
+  - name: file2c
+    path: "/tmp/mgmtC/f2c"
+    content: |
+      i am f2
+    state: exists
+  - name: "@@file3c"
+    path: "/tmp/mgmtC/f3c"
+    content: |
+      i am f3, exported from host C
+    state: exists
+  - name: "@@file4c"
+    path: "/tmp/mgmtC/f4c"
+    content: |
+      i am f4, exported from host C
+    state: exists
+collect:
+- type: file
+  pattern: "/tmp/mgmtC/"
+edges: []

examples/graph4.yaml

@@ -0,0 +1,18 @@
+---
+graph: mygraph
+types:
+  file:
+  - name: file1
+    path: "/tmp/mgmt/f1"
+    content: |
+      i am f1
+    state: exists
+  - name: "@@file3"
+    path: "/tmp/mgmt/f3"
+    content: |
+      i am f3, exported from host A
+    state: exists
+collect:
+- type: file
+  pattern: ''
+edges:

examples/graph5.yaml

@@ -0,0 +1,13 @@
+---
+graph: mygraph
+types:
+  file:
+  - name: file1
+    path: "/tmp/mgmt/f1"
+    content: |
+      i am f1
+    state: exists
+collect:
+- type: file
+  pattern: ''
+edges:

examples/graph6.yaml

@@ -0,0 +1,6 @@
+---
+graph: mygraph
+types:
+  noop:
+  - name: noop1
+edges:

examples/graph7.yaml

@@ -0,0 +1,17 @@
+---
+graph: mygraph
+types:
+  noop:
+  - name: noop1
+  exec:
+  - name: exec1
+    cmd: sleep 10s
+    shell: ''
+    timeout: 0
+    watchcmd: ''
+    watchshell: ''
+    ifcmd: ''
+    ifshell: ''
+    pollint: 0
+    state: present
+edges:

examples/graph8.yaml

@@ -0,0 +1,59 @@
+---
+graph: mygraph
+types:
+  exec:
+  - name: exec1
+    cmd: sleep 10s
+    shell: ''
+    timeout: 0
+    watchcmd: ''
+    watchshell: ''
+    ifcmd: ''
+    ifshell: ''
+    pollint: 0
+    state: present
+  - name: exec2
+    cmd: sleep 10s
+    shell: ''
+    timeout: 0
+    watchcmd: ''
+    watchshell: ''
+    ifcmd: ''
+    ifshell: ''
+    pollint: 0
+    state: present
+  - name: exec3
+    cmd: sleep 10s
+    shell: ''
+    timeout: 0
+    watchcmd: ''
+    watchshell: ''
+    ifcmd: ''
+    ifshell: ''
+    pollint: 0
+    state: present
+  - name: exec4
+    cmd: sleep 10s
+    shell: ''
+    timeout: 0
+    watchcmd: ''
+    watchshell: ''
+    ifcmd: ''
+    ifshell: ''
+    pollint: 0
+    state: present
+edges:
+- name: e1
+  from:
+    type: exec
+    name: exec1
+  to:
+    type: exec
+    name: exec2
+- name: e2
+  from:
+    type: exec
+    name: exec2
+  to:
+    type: exec
+    name: exec3

examples/graph8a.yaml

@@ -0,0 +1,32 @@
+---
+graph: mygraph
+types:
+  exec:
+  - name: exec1
+    cmd: sleep 10s
+    shell: ''
+    timeout: 0
+    watchcmd: ''
+    watchshell: ''
+    ifcmd: ''
+    ifshell: ''
+    pollint: 0
+    state: present
+  - name: exec2
+    cmd: sleep 10s
+    shell: ''
+    timeout: 0
+    watchcmd: ''
+    watchshell: ''
+    ifcmd: ''
+    ifshell: ''
+    pollint: 0
+    state: present
+edges:
+- name: e1
+  from:
+    type: exec
+    name: exec1
+  to:
+    type: exec
+    name: exec2

examples/graph8b.yaml

@@ -0,0 +1,32 @@
+---
+graph: mygraph
+types:
+  exec:
+  - name: exec1
+    cmd: sleep 10s
+    shell: ''
+    timeout: 0
+    watchcmd: ''
+    watchshell: ''
+    ifcmd: 'true'
+    ifshell: ''
+    pollint: 0
+    state: present
+  - name: exec2
+    cmd: sleep 10s
+    shell: ''
+    timeout: 0
+    watchcmd: ''
+    watchshell: ''
+    ifcmd: 'true'
+    ifshell: ''
+    pollint: 0
+    state: present
+edges:
+- name: e1
+  from:
+    type: exec
+    name: exec1
+  to:
+    type: exec
+    name: exec2

examples/graph8c.yaml

@@ -0,0 +1,32 @@
+---
+graph: mygraph
+types:
+  exec:
+  - name: exec1
+    cmd: echo hello from exec1
+    shell: ''
+    timeout: 0
+    watchcmd: sleep 10s
+    watchshell: ''
+    ifcmd: ''
+    ifshell: ''
+    pollint: 0
+    state: present
+  - name: exec2
+    cmd: echo hello from exec2
+    shell: ''
+    timeout: 0
+    watchcmd: sleep 10s
+    watchshell: ''
+    ifcmd: ''
+    ifshell: ''
+    pollint: 0
+    state: present
+edges:
+- name: e1
+  from:
+    type: exec
+    name: exec1
+  to:
+    type: exec
+    name: exec2

examples/graph8d.yaml

@@ -0,0 +1,15 @@
+---
+graph: mygraph
+types:
+  exec:
+  - name: exec1
+    cmd: echo hello from exec1
+    shell: ''
+    timeout: 0
+    watchcmd: sleep 5s
+    watchshell: ''
+    ifcmd: ''
+    ifshell: ''
+    pollint: 0
+    state: present
+edges: []

examples/graph9.yaml

@@ -0,0 +1,77 @@
+---
+graph: mygraph
+comment: simple exec fan in example to demonstrate optimization
+types:
+  exec:
+  - name: exec1
+    cmd: sleep 10s
+    shell: ''
+    timeout: 0
+    watchcmd: ''
+    watchshell: ''
+    ifcmd: ''
+    ifshell: ''
+    pollint: 0
+    state: present
+  - name: exec2
+    cmd: sleep 10s
+    shell: ''
+    timeout: 0
+    watchcmd: ''
+    watchshell: ''
+    ifcmd: ''
+    ifshell: ''
+    pollint: 0
+    state: present
+  - name: exec3
+    cmd: sleep 10s
+    shell: ''
+    timeout: 0
+    watchcmd: ''
+    watchshell: ''
+    ifcmd: ''
+    ifshell: ''
+    pollint: 0
+    state: present
+  - name: exec4
+    cmd: sleep 10s
+    shell: ''
+    timeout: 0
+    watchcmd: ''
+    watchshell: ''
+    ifcmd: ''
+    ifshell: ''
+    pollint: 0
+    state: present
+  - name: exec5
+    cmd: sleep 10s
+    shell: ''
+    timeout: 0
+    watchcmd: ''
+    watchshell: ''
+    ifcmd: ''
+    ifshell: ''
+    pollint: 0
+    state: present
+edges:
+- name: e1
+  from:
+    type: exec
+    name: exec1
+  to:
+    type: exec
+    name: exec5
+- name: e2
+  from:
+    type: exec
+    name: exec2
+  to:
+    type: exec
+    name: exec5
+- name: e3
+  from:
+    type: exec
+    name: exec3
+  to:
+    type: exec
+    name: exec5

examples/purpleidea.service

@@ -0,0 +1,8 @@
+[Unit]
+Description=Fake service for testing
+
+[Service]
+ExecStart=/usr/bin/sleep 8h
+
+[Install]
+WantedBy=multi-user.target

exec.go

@@ -0,0 +1,339 @@
+// Mgmt
+// Copyright (C) 2013-2016+ James Shubin and the project contributors
+// Written by James Shubin <james@shubin.ca> and the project contributors
+//
+// This program is free software: you can redistribute it and/or modify
+// it under the terms of the GNU Affero General Public License as published by
+// the Free Software Foundation, either version 3 of the License, or
+// (at your option) any later version.
+//
+// This program is distributed in the hope that it will be useful,
+// but WITHOUT ANY WARRANTY; without even the implied warranty of
+// MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
+// GNU Affero General Public License for more details.
+//
+// You should have received a copy of the GNU Affero General Public License
+// along with this program.  If not, see <http://www.gnu.org/licenses/>.
+
+package main
+
+import (
+	"bufio"
+	"bytes"
+	"log"
+	"os/exec"
+	"strings"
+)
+
+type ExecType struct {
+	BaseType   `yaml:",inline"`
+	State      string `yaml:"state"`      // state: exists/present?, absent, (undefined?)
+	Cmd        string `yaml:"cmd"`        // the command to run
+	Shell      string `yaml:"shell"`      // the (optional) shell to use to run the cmd
+	Timeout    int    `yaml:"timeout"`    // the cmd timeout in seconds
+	WatchCmd   string `yaml:"watchcmd"`   // the watch command to run
+	WatchShell string `yaml:"watchshell"` // the (optional) shell to use to run the watch cmd
+	IfCmd      string `yaml:"ifcmd"`      // the if command to run
+	IfShell    string `yaml:"ifshell"`    // the (optional) shell to use to run the if cmd
+	PollInt    int    `yaml:"pollint"`    // the poll interval for the ifcmd
+}
+
+func NewExecType(name, cmd, shell string, timeout int, watchcmd, watchshell, ifcmd, ifshell string, pollint int, state string) *ExecType {
+	// FIXME if path = nil, path = name ...
+	return &ExecType{
+		BaseType: BaseType{
+			Name:   name,
+			events: make(chan Event),
+			vertex: nil,
+		},
+		Cmd:        cmd,
+		Shell:      shell,
+		Timeout:    timeout,
+		WatchCmd:   watchcmd,
+		WatchShell: watchshell,
+		IfCmd:      ifcmd,
+		IfShell:    ifshell,
+		PollInt:    pollint,
+		State:      state,
+	}
+}
+
+func (obj *ExecType) GetType() string {
+	return "Exec"
+}
+
+// validate if the params passed in are valid data
+// FIXME: where should this get called ?
+func (obj *ExecType) Validate() bool {
+	if obj.Cmd == "" { // this is the only thing that is really required
+		return false
+	}
+
+	// if we have a watch command, then we don't poll with the if command!
+	if obj.WatchCmd != "" && obj.PollInt > 0 {
+		return false
+	}
+
+	return true
+}
+
+// wraps the scanner output in a channel
+func (obj *ExecType) BufioChanScanner(scanner *bufio.Scanner) (chan string, chan error) {
+	ch, errch := make(chan string), make(chan error)
+	go func() {
+		for scanner.Scan() {
+			ch <- scanner.Text() // blocks here ?
+			if e := scanner.Err(); e != nil {
+				errch <- e // send any misc errors we encounter
+				//break // TODO ?
+			}
+		}
+		close(ch)
+		errch <- scanner.Err() // eof or some err
+		close(errch)
+	}()
+	return ch, errch
+}
+
+// Exec watcher
+func (obj *ExecType) Watch() {
+	if obj.IsWatching() {
+		return
+	}
+	obj.SetWatching(true)
+	defer obj.SetWatching(false)
+
+	var send = false // send event?
+	var exit = false
+	bufioch, errch := make(chan string), make(chan error)
+	//vertex := obj.GetVertex()         // stored with SetVertex
+
+	if obj.WatchCmd != "" {
+		var cmdName string
+		var cmdArgs []string
+		if obj.WatchShell == "" {
+			// call without a shell
+			// FIXME: are there still whitespace splitting issues?
+			split := strings.Fields(obj.WatchCmd)
+			cmdName = split[0]
+			//d, _ := os.Getwd() // TODO: how does this ever error ?
+			//cmdName = path.Join(d, cmdName)
+			cmdArgs = split[1:len(split)]
+		} else {
+			cmdName = obj.Shell // usually bash, or sh
+			cmdArgs = []string{"-c", obj.WatchCmd}
+		}
+		cmd := exec.Command(cmdName, cmdArgs...)
+		//cmd.Dir = "" // look for program in pwd ?
+
+		cmdReader, err := cmd.StdoutPipe()
+		if err != nil {
+			log.Printf("%v[%v]: Error creating StdoutPipe for Cmd: %v", obj.GetType(), obj.GetName(), err)
+			log.Fatal(err) // XXX: how should we handle errors?
+		}
+		scanner := bufio.NewScanner(cmdReader)
+
+		defer cmd.Wait() // XXX: is this necessary?
+		defer func() {
+			// FIXME: without wrapping this in this func it panic's
+			// when running examples/graph8d.yaml
+			cmd.Process.Kill() // TODO: is this necessary?
+		}()
+		if err := cmd.Start(); err != nil {
+			log.Printf("%v[%v]: Error starting Cmd: %v", obj.GetType(), obj.GetName(), err)
+			log.Fatal(err) // XXX: how should we handle errors?
+		}
+
+		bufioch, errch = obj.BufioChanScanner(scanner)
+	}
+
+	for {
+		obj.SetState(typeWatching) // reset
+		select {
+		case text := <-bufioch:
+			obj.SetConvergedState(typeConvergedNil)
+			// each time we get a line of output, we loop!
+			log.Printf("%v[%v]: Watch output: %s", obj.GetType(), obj.GetName(), text)
+			if text != "" {
+				send = true
+			}
+
+		case err := <-errch:
+			obj.SetConvergedState(typeConvergedNil) // XXX ?
+			if err == nil {                         // EOF
+				// FIXME: add an "if watch command ends/crashes"
+				// restart or generate error option
+				log.Printf("%v[%v]: Reached EOF", obj.GetType(), obj.GetName())
+				return
+			}
+			log.Printf("%v[%v]: Error reading input?: %v", obj.GetType(), obj.GetName(), err)
+			log.Fatal(err)
+			// XXX: how should we handle errors?
+
+		case event := <-obj.events:
+			obj.SetConvergedState(typeConvergedNil)
+			if exit, send = obj.ReadEvent(&event); exit {
+				return // exit
+			}
+
+		case _ = <-TimeAfterOrBlock(obj.ctimeout):
+			obj.SetConvergedState(typeConvergedTimeout)
+			obj.converged <- true
+			continue
+		}
+
+		// do all our event sending all together to avoid duplicate msgs
+		if send {
+			send = false
+			// it is okay to invalidate the clean state on poke too
+			obj.isStateOK = false // something made state dirty
+			Process(obj)          // XXX: rename this function
+		}
+	}
+}
+
+// TODO: expand the IfCmd to be a list of commands
+func (obj *ExecType) StateOK() bool {
+
+	// if there is a watch command, but no if command, run based on state
+	if b := obj.isStateOK; obj.WatchCmd != "" && obj.IfCmd == "" {
+		obj.isStateOK = true // reset
+		//if !obj.isStateOK { obj.isStateOK = true; return false }
+		return b
+
+		// if there is no watcher, but there is an onlyif check, run it to see
+	} else if obj.IfCmd != "" { // && obj.WatchCmd == ""
+		// there is a watcher, but there is also an if command
+		//} else if obj.IfCmd != "" && obj.WatchCmd != "" {
+
+		if obj.PollInt > 0 { // && obj.WatchCmd == ""
+			// XXX have the Watch() command output onlyif poll events...
+			// XXX we can optimize by saving those results for returning here
+			// return XXX
+		}
+
+		var cmdName string
+		var cmdArgs []string
+		if obj.IfShell == "" {
+			// call without a shell
+			// FIXME: are there still whitespace splitting issues?
+			split := strings.Fields(obj.IfCmd)
+			cmdName = split[0]
+			//d, _ := os.Getwd() // TODO: how does this ever error ?
+			//cmdName = path.Join(d, cmdName)
+			cmdArgs = split[1:len(split)]
+		} else {
+			cmdName = obj.IfShell // usually bash, or sh
+			cmdArgs = []string{"-c", obj.IfCmd}
+		}
+		err := exec.Command(cmdName, cmdArgs...).Run()
+		if err != nil {
+			// TODO: check exit value
+			return true // don't run
+		}
+		return false // just run
+
+		// if there is no watcher and no onlyif check, assume we should run
+	} else { // if obj.WatchCmd == "" && obj.IfCmd == "" {
+		b := obj.isStateOK
+		obj.isStateOK = true
+		return b // just run if state is dirty
+	}
+}
+
+func (obj *ExecType) Apply() bool {
+	log.Printf("%v[%v]: Apply", obj.GetType(), obj.GetName())
+	var cmdName string
+	var cmdArgs []string
+	if obj.Shell == "" {
+		// call without a shell
+		// FIXME: are there still whitespace splitting issues?
+		// TODO: we could make the split character user selectable...!
+		split := strings.Fields(obj.Cmd)
+		cmdName = split[0]
+		//d, _ := os.Getwd() // TODO: how does this ever error ?
+		//cmdName = path.Join(d, cmdName)
+		cmdArgs = split[1:len(split)]
+	} else {
+		cmdName = obj.Shell // usually bash, or sh
+		cmdArgs = []string{"-c", obj.Cmd}
+	}
+	cmd := exec.Command(cmdName, cmdArgs...)
+	//cmd.Dir = "" // look for program in pwd ?
+	var out bytes.Buffer
+	cmd.Stdout = &out
+
+	if err := cmd.Start(); err != nil {
+		log.Printf("%v[%v]: Error starting Cmd: %v", obj.GetType(), obj.GetName(), err)
+		return false
+	}
+
+	timeout := obj.Timeout
+	if timeout == 0 { // zero timeout means no timer, so disable it
+		timeout = -1
+	}
+	done := make(chan error)
+	go func() { done <- cmd.Wait() }()
+
+	select {
+	case err := <-done:
+		if err != nil {
+			log.Printf("%v[%v]: Error waiting for Cmd: %v", obj.GetType(), obj.GetName(), err)
+			return false
+		}
+
+	case <-TimeAfterOrBlock(timeout):
+		log.Printf("%v[%v]: Timeout waiting for Cmd", obj.GetType(), obj.GetName())
+		//cmd.Process.Kill() // TODO: is this necessary?
+		return false
+	}
+
+	// TODO: if we printed the stdout while the command is running, this
+	// would be nice, but it would require terminal log output that doesn't
+	// interleave all the parallel parts which would mix it all up...
+	if s := out.String(); s == "" {
+		log.Printf("Exec[%v]: Command output is empty!", obj.Name)
+	} else {
+		log.Printf("Exec[%v]: Command output is:", obj.Name)
+		log.Printf(out.String())
+	}
+	// XXX: return based on exit value!!
+	return true
+}
+
+func (obj *ExecType) Compare(typ Type) bool {
+	switch typ.(type) {
+	case *ExecType:
+		typ := typ.(*ExecType)
+		if obj.Name != typ.Name {
+			return false
+		}
+		if obj.Cmd != typ.Cmd {
+			return false
+		}
+		if obj.Shell != typ.Shell {
+			return false
+		}
+		if obj.Timeout != typ.Timeout {
+			return false
+		}
+		if obj.WatchCmd != typ.WatchCmd {
+			return false
+		}
+		if obj.WatchShell != typ.WatchShell {
+			return false
+		}
+		if obj.IfCmd != typ.IfCmd {
+			return false
+		}
+		if obj.PollInt != typ.PollInt {
+			return false
+		}
+		if obj.State != typ.State {
+			return false
+		}
+	default:
+		return false
+	}
+	return true
+}

file.go

@@ -1,5 +1,5 @@
 // Mgmt
-// Copyright (C) 2013-2015+ James Shubin and the project contributors
+// Copyright (C) 2013-2016+ James Shubin and the project contributors
 // Written by James Shubin <james@shubin.ca> and the project contributors
 //
 // This program is free software: you can redistribute it and/or modify
@@ -18,10 +18,8 @@
 package main
 
 import (
-	"code.google.com/p/go-uuid/uuid"
 	"crypto/sha256"
 	"encoding/hex"
-	"fmt"
 	"gopkg.in/fsnotify.v1"
 	//"github.com/go-fsnotify/fsnotify" // git master of "gopkg.in/fsnotify.v1"
 	"io"
@@ -34,39 +32,82 @@ import (
 )
 
 type FileType struct {
-	uuid      string
-	Type      string      // always "file"
-	Name      string      // name variable
-	Events    chan string // FIXME: eventually a struct for the event?
-	Path      string      // path variable (should default to name)
-	Content   string
-	State     string // state: exists/present?, absent, (undefined?)
+	BaseType  `yaml:",inline"`
+	Path      string `yaml:"path"` // path variable (should default to name)
+	Dirname   string `yaml:"dirname"`
+	Basename  string `yaml:"basename"`
+	Content   string `yaml:"content"`
+	State     string `yaml:"state"` // state: exists/present?, absent, (undefined?)
 	sha256sum string
 }
 
-func NewFileType(name, path, content, state string) *FileType {
+func NewFileType(name, path, dirname, basename, content, state string) *FileType {
 	// FIXME if path = nil, path = name ...
 	return &FileType{
-		uuid:      uuid.New(),
-		Type:      "file",
-		Name:      name,
-		Events:    make(chan string, 1), // XXX: chan size?
+		BaseType: BaseType{
+			Name:   name,
+			events: make(chan Event),
+			vertex: nil,
+		},
 		Path:      path,
+		Dirname:   dirname,
+		Basename:  basename,
 		Content:   content,
 		State:     state,
 		sha256sum: "",
 	}
 }
 
+func (obj *FileType) GetType() string {
+	return "File"
+}
+
+// validate if the params passed in are valid data
+func (obj *FileType) Validate() bool {
+	if obj.Dirname != "" {
+		// must end with /
+		if obj.Dirname[len(obj.Dirname)-1:] != "/" {
+			return false
+		}
+	}
+	if obj.Basename != "" {
+		// must not start with /
+		if obj.Basename[0:1] == "/" {
+			return false
+		}
+	}
+	return true
+}
+
+func (obj *FileType) GetPath() string {
+	d := Dirname(obj.Path)
+	b := Basename(obj.Path)
+	if !obj.Validate() || (obj.Dirname == "" && obj.Basename == "") {
+		return obj.Path
+	} else if obj.Dirname == "" {
+		return d + obj.Basename
+	} else if obj.Basename == "" {
+		return obj.Dirname + b
+	} else { // if obj.dirname != "" && obj.basename != "" {
+		return obj.Dirname + obj.Basename
+	}
+}
+
 // File watcher for files and directories
 // Modify with caution, probably important to write some test cases first!
-func (obj FileType) Watch(v *Vertex) {
-	// obj.Path: file or directory
-	//var recursive bool = false
-	//var isdir = (obj.Path[len(obj.Path)-1:] == "/") // dirs have trailing slashes
-	//fmt.Printf("IsDirectory: %v\n", isdir)
+// obj.GetPath(): file or directory
+func (obj *FileType) Watch() {
+	if obj.IsWatching() {
+		return
+	}
+	obj.SetWatching(true)
+	defer obj.SetWatching(false)
 
-	var safename = path.Clean(obj.Path) // no trailing slash
+	//var recursive bool = false
+	//var isdir = (obj.GetPath()[len(obj.GetPath())-1:] == "/") // dirs have trailing slashes
+	//log.Printf("IsDirectory: %v", isdir)
+	//vertex := obj.GetVertex()         // stored with SetVertex
+	var safename = path.Clean(obj.GetPath()) // no trailing slash
 
 	watcher, err := fsnotify.NewWatcher()
 	if err != nil {
@@ -77,90 +118,84 @@ func (obj FileType) Watch(v *Vertex) {
 	patharray := PathSplit(safename) // tokenize the path
 	var index = len(patharray)       // starting index
 	var current string               // current "watcher" location
-	var delta_depth int              // depth delta between watcher and event
+	var deltaDepth int               // depth delta between watcher and event
 	var send = false                 // send event?
-	var extraCheck = false
+	var exit = false
+	var dirty = false
 
 	for {
 		current = strings.Join(patharray[0:index], "/")
 		if current == "" { // the empty string top is the root dir ("/")
 			current = "/"
 		}
-		log.Printf("Watching: %v\n", current) // attempting to watch...
-
+		if DEBUG {
+			log.Printf("File[%v]: Watching: %v", obj.GetName(), current) // attempting to watch...
+		}
 		// initialize in the loop so that we can reset on rm-ed handles
 		err = watcher.Add(current)
 		if err != nil {
+			if DEBUG {
+				log.Printf("File[%v]: watcher.Add(%v): Error: %v", obj.GetName(), current, err)
+			}
 			if err == syscall.ENOENT {
 				index-- // usually not found, move up one dir
 			} else if err == syscall.ENOSPC {
-				// XXX: i sometimes see: no space left on device
-				// XXX: why causes this to happen ?
-				log.Printf("Strange file[%v] error: %+v\n", obj.Name, err.Error) // 0x408da0
+				// XXX: occasionally: no space left on device,
+				// XXX: probably due to lack of inotify watches
+				log.Printf("Lack of watches for file[%v] error: %+v", obj.Name, err.Error) // 0x408da0
 				log.Fatal(err)
 			} else {
-				log.Printf("Unknown file[%v] error:\n", obj.Name)
+				log.Printf("Unknown file[%v] error:", obj.Name)
 				log.Fatal(err)
 			}
 			index = int(math.Max(1, float64(index)))
 			continue
 		}
 
-		// XXX: check state after inotify started
-		// SMALL RACE: after we terminate watch, till when it's started
-		// something could have gotten created/changed/etc... right?
-		if extraCheck {
-			extraCheck = false
-			// XXX
-			//if exists ... {
-			//	send signal
-			//	continue
-			//	change index? i don't think so. be thorough and check
-			//}
-		}
-
+		obj.SetState(typeWatching) // reset
 		select {
 		case event := <-watcher.Events:
-			// the deeper you go, the bigger the delta_depth is...
+			if DEBUG {
+				log.Printf("File[%v]: Watch(%v), Event(%v): %v", obj.GetName(), current, event.Name, event.Op)
+			}
+			obj.SetConvergedState(typeConvergedNil) // XXX: technically i can detect if the event is erroneous or not first
+			// the deeper you go, the bigger the deltaDepth is...
 			// this is the difference between what we're watching,
 			// and the event... doesn't mean we can't watch deeper
 			if current == event.Name {
-				delta_depth = 0 // i was watching what i was looking for
+				deltaDepth = 0 // i was watching what i was looking for
 
 			} else if HasPathPrefix(event.Name, current) {
-				delta_depth = len(PathSplit(current)) - len(PathSplit(event.Name)) // -1 or less
+				deltaDepth = len(PathSplit(current)) - len(PathSplit(event.Name)) // -1 or less
 
 			} else if HasPathPrefix(current, event.Name) {
-				delta_depth = len(PathSplit(event.Name)) - len(PathSplit(current)) // +1 or more
+				deltaDepth = len(PathSplit(event.Name)) - len(PathSplit(current)) // +1 or more
 
 			} else {
-				// XXX multiple watchers receive each others events
+				// TODO different watchers get each others events!
 				// https://github.com/go-fsnotify/fsnotify/issues/95
 				// this happened with two values such as:
 				// event.Name: /tmp/mgmt/f3 and current: /tmp/mgmt/f2
-				// are the different watchers getting each others events??
-				//log.Printf("The delta depth is NaN...\n")
-				//log.Printf("Value of event.Name is: %v\n", event.Name)
-				//log.Printf("........    current is: %v\n", current)
-				//log.Fatal("The delta depth is NaN!")
 				continue
 			}
-			//log.Printf("The delta depth is: %v\n", delta_depth)
+			//log.Printf("The delta depth is: %v", deltaDepth)
 
 			// if we have what we wanted, awesome, send an event...
 			if event.Name == safename {
 				//log.Println("Event!")
+				// FIXME: should all these below cases trigger?
 				send = true
+				dirty = true
 
 				// file removed, move the watch upwards
-				if delta_depth >= 0 && (event.Op&fsnotify.Remove == fsnotify.Remove) {
+				if deltaDepth >= 0 && (event.Op&fsnotify.Remove == fsnotify.Remove) {
 					//log.Println("Removal!")
 					watcher.Remove(current)
 					index--
 				}
 
 				// we must be a parent watcher, so descend in
-				if delta_depth < 0 {
+				if deltaDepth < 0 {
 					watcher.Remove(current)
 					index++
 				}
@@ -169,13 +204,18 @@ func (obj FileType) Watch(v *Vertex) {
 			} else if HasPathPrefix(safename, event.Name) {
 				//log.Println("Above!")
 
-				if delta_depth >= 0 && (event.Op&fsnotify.Remove == fsnotify.Remove) {
+				if deltaDepth >= 0 && (event.Op&fsnotify.Remove == fsnotify.Remove) {
 					log.Println("Removal!")
 					watcher.Remove(current)
 					index--
 				}
 
-				if delta_depth < 0 {
+				if deltaDepth < 0 {
+					log.Println("Parent!")
+					if PathPrefixDelta(safename, event.Name) == 1 { // we're the parent dir
+						send = true
+						dirty = true
+					}
 					watcher.Remove(current)
 					index++
 				}
@@ -184,37 +224,42 @@ func (obj FileType) Watch(v *Vertex) {
 			} else if HasPathPrefix(event.Name, safename) {
 				//log.Println("Event2!")
 				send = true
+				dirty = true
 			}
 
 		case err := <-watcher.Errors:
+			obj.SetConvergedState(typeConvergedNil) // XXX ?
 			log.Println("error:", err)
 			log.Fatal(err)
-			v.Events <- fmt.Sprintf("file: %v", "error")
+			//obj.events <- fmt.Sprintf("file: %v", "error") // XXX: how should we handle errors?
 
-		case exit := <-obj.Events:
-			if exit == "exit" {
-				return
-			} else {
-				log.Fatal("Unknown event: %v\n", exit)
+		case event := <-obj.events:
+			obj.SetConvergedState(typeConvergedNil)
+			if exit, send = obj.ReadEvent(&event); exit {
+				return // exit
 			}
+			//dirty = false // these events don't invalidate state
+
+		case _ = <-TimeAfterOrBlock(obj.ctimeout):
+			obj.SetConvergedState(typeConvergedTimeout)
+			obj.converged <- true
+			continue
 		}
 
 		// do all our event sending all together to avoid duplicate msgs
 		if send {
 			send = false
-			//log.Println("Sending event!")
-			//v.Events <- fmt.Sprintf("file(%v): %v", obj.Path, event.Op)
-			v.Events <- fmt.Sprintf("file(%v): %v", obj.Path, "event!") // FIXME: use struct
+			// only invalid state on certain types of events
+			if dirty {
+				dirty = false
+				obj.isStateOK = false // something made state dirty
+			}
+			Process(obj) // XXX: rename this function
 		}
 	}
 }
 
-func (obj FileType) Exit() bool {
-	obj.Events <- "exit"
-	return true
-}
-
-func (obj FileType) HashSHA256fromContent() string {
+func (obj *FileType) HashSHA256fromContent() string {
 	if obj.sha256sum != "" { // return if already computed
 		return obj.sha256sum
 	}
@@ -225,11 +270,16 @@ func (obj FileType) HashSHA256fromContent() string {
 	return obj.sha256sum
 }
 
-func (obj FileType) StateOK() bool {
-	if _, err := os.Stat(obj.Path); os.IsNotExist(err) {
+// FIXME: add the obj.CleanState() calls all over the true returns!
+func (obj *FileType) StateOK() bool {
+	if obj.isStateOK { // cache the state
+		return true
+	}
+
+	if _, err := os.Stat(obj.GetPath()); os.IsNotExist(err) {
 		// no such file or directory
 		if obj.State == "absent" {
-			return true // missing file should be missing, phew :)
+			return obj.CleanState() // missing file should be missing, phew :)
 		} else {
 			// state invalid, skip expensive checksums
 			return false
@@ -238,15 +288,15 @@ func (obj FileType) StateOK() bool {
 
 	// TODO: add file mode check here...
 
-	if PathIsDir(obj.Path) {
+	if PathIsDir(obj.GetPath()) {
 		return obj.StateOKDir()
 	} else {
 		return obj.StateOKFile()
 	}
 }
 
-func (obj FileType) StateOKFile() bool {
-	if PathIsDir(obj.Path) {
+func (obj *FileType) StateOKFile() bool {
+	if PathIsDir(obj.GetPath()) {
 		log.Fatal("This should only be called on a File type.")
 	}
 
@@ -254,7 +304,7 @@ func (obj FileType) StateOKFile() bool {
 
 	hash := sha256.New()
 
-	f, err := os.Open(obj.Path)
+	f, err := os.Open(obj.GetPath())
 	if err != nil {
 		//log.Fatal(err)
 		return false
@@ -267,7 +317,7 @@ func (obj FileType) StateOKFile() bool {
 	}
 
 	sha256sum := hex.EncodeToString(hash.Sum(nil))
-	//fmt.Printf("sha256sum: %v\n", sha256sum)
+	//log.Printf("sha256sum: %v", sha256sum)
 
 	if obj.HashSHA256fromContent() == sha256sum {
 		return true
@@ -276,8 +326,8 @@ func (obj FileType) StateOKFile() bool {
 	return false
 }
 
-func (obj FileType) StateOKDir() bool {
-	if !PathIsDir(obj.Path) {
+func (obj *FileType) StateOKDir() bool {
+	if !PathIsDir(obj.GetPath()) {
 		log.Fatal("This should only be called on a Dir type.")
 	}
 
@@ -286,33 +336,33 @@ func (obj FileType) StateOKDir() bool {
 	return false
 }
 
-func (obj FileType) Apply() bool {
-	fmt.Printf("Apply->%v[%v]\n", obj.Type, obj.Name)
+func (obj *FileType) Apply() bool {
+	log.Printf("%v[%v]: Apply", obj.GetType(), obj.GetName())
 
-	if PathIsDir(obj.Path) {
+	if PathIsDir(obj.GetPath()) {
 		return obj.ApplyDir()
 	} else {
 		return obj.ApplyFile()
 	}
 }
 
-func (obj FileType) ApplyFile() bool {
+func (obj *FileType) ApplyFile() bool {
 
-	if PathIsDir(obj.Path) {
+	if PathIsDir(obj.GetPath()) {
 		log.Fatal("This should only be called on a File type.")
 	}
 
 	if obj.State == "absent" {
-		log.Printf("About to remove: %v\n", obj.Path)
-		err := os.Remove(obj.Path)
+		log.Printf("About to remove: %v", obj.GetPath())
+		err := os.Remove(obj.GetPath())
 		if err != nil {
 			return false
 		}
 		return true
 	}
 
-	//fmt.Println("writing: " + filename)
-	f, err := os.Create(obj.Path)
+	//log.Println("writing: " + filename)
+	f, err := os.Create(obj.GetPath())
 	if err != nil {
 		log.Println("error:", err)
 		return false
@@ -328,8 +378,8 @@ func (obj FileType) ApplyFile() bool {
 	return true
 }
 
-func (obj FileType) ApplyDir() bool {
-	if !PathIsDir(obj.Path) {
+func (obj *FileType) ApplyDir() bool {
+	if !PathIsDir(obj.GetPath()) {
 		log.Fatal("This should only be called on a Dir type.")
 	}
 
@@ -337,3 +387,25 @@ func (obj FileType) ApplyDir() bool {
 	log.Fatal("Not implemented!")
 	return true
 }
+
+func (obj *FileType) Compare(typ Type) bool {
+	switch typ.(type) {
+	case *FileType:
+		typ := typ.(*FileType)
+		if obj.Name != typ.Name {
+			return false
+		}
+		if obj.GetPath() != typ.Path {
+			return false
+		}
+		if obj.Content != typ.Content {
+			return false
+		}
+		if obj.State != typ.State {
+			return false
+		}
+	default:
+		return false
+	}
+	return true
+}

main.go

@@ -1,5 +1,5 @@
 // Mgmt
-// Copyright (C) 2013-2015+ James Shubin and the project contributors
+// Copyright (C) 2013-2016+ James Shubin and the project contributors
 // Written by James Shubin <james@shubin.ca> and the project contributors
 //
 // This program is free software: you can redistribute it and/or modify
@@ -18,7 +18,6 @@
 package main
 
 import (
-	"fmt"
 	"github.com/codegangsta/cli"
 	"log"
 	"os"
@@ -26,12 +25,13 @@ import (
 	"sync"
 	"syscall"
 	"time"
+	//etcd_context "github.com/coreos/etcd/Godeps/_workspace/src/golang.org/x/net/context"
 )
 
 // set at compile time
 var (
-	version string
 	program string
+	version string
 )
 
 const (
@@ -48,7 +48,6 @@ func waitForSignal(exit chan bool) {
 	select {
 	case e := <-signals: // any signal will do
 		if e == os.Interrupt {
-			fmt.Println() // put ^C char from terminal on its own line
 			log.Println("Interrupted by ^C")
 		} else {
 			log.Println("Interrupted by signal")
@@ -59,54 +58,149 @@ func waitForSignal(exit chan bool) {
 }
 
 func run(c *cli.Context) {
-	var start int64 = time.Now().UnixNano()
+	var start = time.Now().UnixNano()
 	var wg sync.WaitGroup
-	exit := make(chan bool) // exit signal
-	log.Printf("This is: %v, version: %v\n", program, version)
+	exit := make(chan bool)      // exit signal
+	converged := make(chan bool) // converged signal
+	log.Printf("This is: %v, version: %v", program, version)
+	log.Printf("Main: Start: %v", start)
+	G := NewGraph("Graph") // give graph a default name
 
-	// exit after `exittime` seconds for no reason at all...
-	if i := c.Int("exittime"); i > 0 {
+	// exit after `max-runtime` seconds for no reason at all...
+	if i := c.Int("max-runtime"); i > 0 {
 		go func() {
 			time.Sleep(time.Duration(i) * time.Second)
 			exit <- true
 		}()
 	}
 
-	// build the graph from a config file
-	G := GraphFromConfig(c.String("file"))
-	log.Printf("Graph: %v\n", G) // show graph
+	// initial etcd peer endpoint
+	seed := c.String("seed")
+	if seed == "" {
+		// XXX: start up etcd server, others will join me!
+		seed = "http://127.0.0.1:2379" // thus we use the local server!
+	}
+	// then, connect to `seed` as a client
 
-	log.Printf("Start: %v\n", start)
+	// FIXME: validate seed, or wait for it to fail in etcd init?
 
-	for x := range G.GetVerticesChan() { // XXX ?
-		log.Printf("Main->Starting[%v]\n", x.Name)
-
-		wg.Add(1)
-		// must pass in value to avoid races...
-		// see: https://ttboj.wordpress.com/2015/07/27/golang-parallelism-issues-causing-too-many-open-files-error/
-		go func(v *Vertex) {
-			defer wg.Done()
-			v.Start()
-			log.Printf("Main->Finish[%v]\n", v.Name)
-		}(x)
-
-		// generate a startup "poke" so that an initial check happens
-		go func(v *Vertex) {
-			v.Events <- fmt.Sprintf("Startup(%v)", v.Name)
-		}(x)
+	// etcd
+	etcdO := &EtcdWObject{
+		seed:      seed,
+		ctimeout:  c.Int("converged-timeout"),
+		converged: converged,
 	}
 
-	log.Println("Running...")
+	hostname := c.String("hostname")
+	if hostname == "" {
+		hostname, _ = os.Hostname() // etcd watch key // XXX: this is not the correct key name this is the set key name... WOOPS
+	}
+	go func() {
+		startchan := make(chan struct{}) // start signal
+		go func() { startchan <- struct{}{} }()
+		file := c.String("file")
+		configchan := make(chan bool)
+		if !c.Bool("no-watch") {
+			configchan = ConfigWatch(file)
+		}
+		log.Printf("Etcd: Starting...")
+		etcdchan := etcdO.EtcdWatch()
+		first := true // first loop or not
+		for {
+			select {
+			case _ = <-startchan: // kick the loop once at start
+				// pass
+			case msg := <-etcdchan:
+				switch msg {
+				// some types of messages we ignore...
+				case etcdFoo, etcdBar:
+					continue
+				// while others passthrough and cause a compile!
+				case etcdStart, etcdEvent:
+					// pass
+				default:
+					log.Fatal("Etcd: Unhandled message: ", msg)
+				}
+			case msg := <-configchan:
+				if c.Bool("no-watch") || !msg {
+					continue // not ready to read config
+				}
+				//case compile_event: XXX
+			}
+
+			config := ParseConfigFromFile(file)
+			if config == nil {
+				log.Printf("Config parse failure")
+				continue
+			}
+
+			// run graph vertex LOCK...
+			if !first { // XXX: we can flatten this check out I think
+				log.Printf("State: %v -> %v", G.SetState(graphPausing), G.GetState())
+				G.Pause() // sync
+				log.Printf("State: %v -> %v", G.SetState(graphPaused), G.GetState())
+			}
+
+			// build the graph from a config file
+			// build the graph on events (eg: from etcd)
+			if !UpdateGraphFromConfig(config, hostname, G, etcdO) {
+				log.Fatal("Config: We borked the graph.") // XXX
+			}
+			log.Printf("Graph: %v", G) // show graph
+			err := G.ExecGraphviz(c.String("graphviz-filter"), c.String("graphviz"))
+			if err != nil {
+				log.Printf("Graphviz: %v", err)
+			} else {
+				log.Printf("Graphviz: Successfully generated graph!")
+			}
+			G.SetVertex()
+			G.SetConvergedCallback(c.Int("converged-timeout"), converged)
+			// G.Start(...) needs to be synchronous or wait,
+			// because if half of the nodes are started and
+			// some are not ready yet and the EtcdWatch
+			// loops, we'll cause G.Pause(...) before we
+			// even got going, thus causing nil pointer errors
+			log.Printf("State: %v -> %v", G.SetState(graphStarting), G.GetState())
+			G.Start(&wg, first) // sync
+			log.Printf("State: %v -> %v", G.SetState(graphStarted), G.GetState())
+			first = false
+		}
+	}()
+
+	if i := c.Int("converged-timeout"); i >= 0 {
+		go func() {
+		ConvergedLoop:
+			for {
+				<-converged // when anyone says they have converged
+
+				if etcdO.GetConvergedState() != etcdConvergedTimeout {
+					continue
+				}
+				for v := range G.GetVerticesChan() {
+					if v.Type.GetConvergedState() != typeConvergedTimeout {
+						continue ConvergedLoop
+					}
+				}
+
+				// if all have converged, exit
+				log.Printf("Converged for %d seconds, exiting!", i)
+				exit <- true
+				for {
+					<-converged
+				} // unblock/drain
+				//return
+			}
+		}()
+	}
+
+	log.Println("Main: Running...")
 
 	waitForSignal(exit) // pass in exit channel to watch
 
 	G.Exit() // tell all the children to exit
 
 	if DEBUG {
-		for i := range G.GetVerticesChan() {
-			fmt.Printf("Vertex: %v\n", i)
-		}
-		fmt.Printf("Graph: %v\n", G)
+		log.Printf("Graph: %v", G)
 	}
 
 	wg.Wait() // wait for primary go routines to exit
@@ -116,6 +210,13 @@ func run(c *cli.Context) {
 }
 
 func main() {
+	//if DEBUG {
+	log.SetFlags(log.LstdFlags | log.Lshortfile)
+	//}
+	log.SetFlags(log.Flags() - log.Ldate) // remove the date for now
+	if program == "" || version == "" {
+		log.Fatal("Program was not compiled correctly. Please see Makefile.")
+	}
 	app := cli.NewApp()
 	app.Name = program
 	app.Usage = "next generation config management"
@@ -134,8 +235,44 @@ func main() {
 					Value: "",
 					Usage: "graph definition to run",
 				},
+				cli.BoolFlag{
+					Name:  "no-watch",
+					Usage: "do not update graph on watched graph definition file changes",
+				},
+				cli.StringFlag{
+					Name:  "code, c",
+					Value: "",
+					Usage: "code definition to run",
+				},
+				cli.StringFlag{
+					Name:  "graphviz, g",
+					Value: "",
+					Usage: "output file for graphviz data",
+				},
+				cli.StringFlag{
+					Name:  "graphviz-filter, gf",
+					Value: "dot", // directed graph default
+					Usage: "graphviz filter to use",
+				},
+				// useful for testing multiple instances on same machine
+				cli.StringFlag{
+					Name:  "hostname",
+					Value: "",
+					Usage: "hostname to use",
+				},
+				// if empty, it will startup a new server
+				cli.StringFlag{
+					Name:  "seed, s",
+					Value: "",
+					Usage: "default etc peer endpoint",
+				},
 				cli.IntFlag{
-					Name:  "exittime",
+					Name:  "converged-timeout, t",
+					Value: -1,
+					Usage: "exit after approximately this many seconds in a converged state",
+				},
+				cli.IntFlag{
+					Name:  "max-runtime",
 					Value: 0,
 					Usage: "exit after a maximum of approximately this many seconds",
 				},

main_test.go

@@ -1,5 +1,5 @@
 // Mgmt
-// Copyright (C) 2013-2015+ James Shubin and the project contributors
+// Copyright (C) 2013-2016+ James Shubin and the project contributors
 // Written by James Shubin <james@shubin.ca> and the project contributors
 //
 // This program is free software: you can redistribute it and/or modify

mgmt.spec.in

@@ -0,0 +1,63 @@
+%global project_version __VERSION__
+%define debug_package %{nil}
+
+Name:		mgmt
+Version:	__VERSION__
+Release:	__RELEASE__
+Summary:	A next generation config management prototype!
+License:	AGPLv3+
+URL:		https://github.com/purpleidea/mgmt
+Source0:	https://dl.fedoraproject.org/pub/alt/purpleidea/mgmt/SOURCES/mgmt-%{project_version}.tar.bz2
+
+# graphviz should really be a "suggests", since technically it's optional
+Requires: graphviz
+
+BuildRequires:	golang
+BuildRequires:	golang-googlecode-tools-stringer
+BuildRequires:	git-core
+BuildRequires:	mercurial
+
+%description
+A next generation config management prototype!
+
+%prep
+%setup
+
+%build
+# FIXME: in the future, these could be vendor-ed in
+mkdir -p vendor/
+export GOPATH=`pwd`/vendor/
+go get github.com/coreos/etcd/client
+go get gopkg.in/yaml.v2
+go get gopkg.in/fsnotify.v1
+go get github.com/codegangsta/cli
+go get github.com/coreos/go-systemd/dbus
+go get github.com/coreos/go-systemd/util
+make build
+
+%install
+rm -rf %{buildroot}
+# _datadir is typically /usr/share/
+install -d -m 0755 %{buildroot}/%{_datadir}/mgmt/
+cp -a AUTHORS COPYING COPYRIGHT DOCUMENTATION.md README.md THANKS examples/ %{buildroot}/%{_datadir}/mgmt/
+
+# install the binary
+mkdir -p %{buildroot}/%{_bindir}
+install -m 0755 mgmt %{buildroot}/%{_bindir}/mgmt
+
+# profile.d bash completion
+mkdir -p %{buildroot}%{_sysconfdir}/profile.d
+install misc/mgmt.bashrc -m 0755 %{buildroot}%{_sysconfdir}/profile.d/mgmt.sh
+
+# etc dir
+mkdir -p %{buildroot}%{_sysconfdir}/mgmt/
+install -m 0644 misc/mgmt.conf.example %{buildroot}%{_sysconfdir}/mgmt/mgmt.conf
+
+%files
+%attr(0755, root, root) %{_sysconfdir}/profile.d/mgmt.sh
+%{_datadir}/mgmt/*
+%{_bindir}/mgmt
+%{_sysconfdir}/mgmt/*
+
+# this changelog is auto-generated by git log
+%changelog

misc.go

@@ -1,5 +1,5 @@
 // Mgmt
-// Copyright (C) 2013-2015+ James Shubin and the project contributors
+// Copyright (C) 2013-2016+ James Shubin and the project contributors
 // Written by James Shubin <james@shubin.ca> and the project contributors
 //
 // This program is free software: you can redistribute it and/or modify
@@ -18,16 +18,31 @@
 package main
 
 import (
+	"bytes"
+	"encoding/base64"
+	"encoding/gob"
 	"path"
 	"strings"
+	"time"
 )
 
 // Similar to the GNU dirname command
 func Dirname(p string) string {
+	if p == "/" {
+		return ""
+	}
 	d, _ := path.Split(path.Clean(p))
 	return d
 }
 
+func Basename(p string) string {
+	_, b := path.Split(path.Clean(p))
+	if p[len(p)-1:] == "/" { // don't loose the tail slash
+		b += "/"
+	}
+	return b
+}
+
 // Split a path into an array of tokens excluding any trailing empty tokens
 func PathSplit(p string) []string {
 	return strings.Split(path.Clean(p), "/")
@@ -52,6 +67,56 @@ func HasPathPrefix(p, prefix string) bool {
 	return true
 }
 
+// Delta of path prefix, tells you how many path tokens different the prefix is
+func PathPrefixDelta(p, prefix string) int {
+
+	if !HasPathPrefix(p, prefix) {
+		return -1
+	}
+	patharray := PathSplit(p)
+	prefixarray := PathSplit(prefix)
+	return len(patharray) - len(prefixarray)
+}
+
 func PathIsDir(p string) bool {
 	return p[len(p)-1:] == "/" // a dir has a trailing slash in this context
 }
+
+// encode an object as base 64, serialize and then base64 encode
+func ObjToB64(obj interface{}) (string, bool) {
+	b := bytes.Buffer{}
+	e := gob.NewEncoder(&b)
+	err := e.Encode(obj)
+	if err != nil {
+		//log.Println("Gob failed to Encode: ", err)
+		return "", false
+	}
+	return base64.StdEncoding.EncodeToString(b.Bytes()), true
+}
+
+// TODO: is it possible to somehow generically just return the obj?
+// decode an object into the waiting obj which you pass a reference to
+func B64ToObj(str string, obj interface{}) bool {
+	bb, err := base64.StdEncoding.DecodeString(str)
+	if err != nil {
+		//log.Println("Base64 failed to Decode: ", err)
+		return false
+	}
+	b := bytes.NewBuffer(bb)
+	d := gob.NewDecoder(b)
+	err = d.Decode(obj)
+	if err != nil {
+		//log.Println("Gob failed to Decode: ", err)
+		return false
+	}
+	return true
+}
+
+// special version of time.After that blocks when given a negative integer
+// when used in a case statement, the timer restarts on each select call to it
+func TimeAfterOrBlock(t int) <-chan time.Time {
+	if t < 0 {
+		return make(chan time.Time) // blocks forever
+	}
+	return time.After(time.Duration(t) * time.Second)
+}

misc/centos-ci.py

@@ -0,0 +1,72 @@
+#!/usr/bin/python
+
+# modified from:
+# https://github.com/kbsingh/centos-ci-scripts/blob/master/build_python_script.py
+# usage: centos-ci.py giturl [branch [commands]]
+
+import os
+import sys
+import json
+import urllib
+import subprocess
+
+# static argv to be used if running script inline
+argv = [
+	#'https://github.com/purpleidea/mgmt', # giturl
+	#'master',
+	#'make test',
+]
+argv.insert(0, '') # add a fake argv[0]
+url_base = 'http://admin.ci.centos.org:8080'
+apikey = '' # put api key here if running inline
+if apikey == '':
+	apikey = os.environ.get('DUFFY_API_KEY')
+if apikey is None or apikey == '':
+	apikey = open('duffy.key', 'r').read().strip()
+ver = '7'
+arch = 'x86_64'
+count = 1
+
+if len(argv) <= 1: argv = sys.argv # use system argv because ours is empty
+if len(argv) <= 1:
+	print 'Not enough arguments supplied!'
+	sys.exit(1)
+
+git_url = argv[1]
+branch = 'master'
+if len(argv) > 2: branch = argv[2]
+folder = os.path.basename(git_url)	# should be project name
+run = 'make vtest' # the omv vtest cmd is a good option to run from this target
+if len(argv) > 3: run = ' '.join(argv[3:])
+
+get_nodes_url = "%s/Node/get?key=%s&ver=%s&arch=%s&i_count=%s" % (url_base, apikey, ver, arch, count)
+data = json.loads(urllib.urlopen(get_nodes_url).read()) # request host(s)
+hosts = data['hosts']
+ssid = data['ssid']
+done_nodes_url = "%s/Node/done?key=%s&ssid=%s" % (url_base, apikey, ssid)
+
+host = hosts[0]
+ssh = "ssh -tt -o UserKnownHostsFile=/dev/null -o StrictHostKeyChecking=no -o SendEnv=JENKINS_URL root@%s" % host
+yum = 'yum -y install git wget tree psmisc'
+omv = 'wget https://github.com/purpleidea/oh-my-vagrant/raw/master/extras/install-omv.sh && chmod u+x install-omv.sh && ./install-omv.sh && wget https://github.com/purpleidea/mgmt/raw/master/misc/make-path.sh && chmod u+x make-path.sh && ./make-path.sh'
+cmd = "%s '%s && %s'" % (ssh, yum, omv) # setup
+print cmd
+r = subprocess.call(cmd, shell=True)
+if r != 0:
+	# NOTE: we don't clean up the host here, so that it can be inspected!
+	print "Error configuring omv on: %s" % host
+	sys.exit(r)
+
+# the second ssh call will run with the omv /etc/profile.d/ script loaded
+git = "git clone --recursive %s %s && cd %s && git checkout %s" % (git_url, folder, folder, branch)
+cmd = "%s 'export JENKINS_URL=%s && %s && %s'" % (ssh, os.getenv('JENKINS_URL', ''), git, run) # run
+print cmd
+r = subprocess.call(cmd, shell=True)
+if r != 0:
+	print "Error running job on: %s" % host
+
+output = urllib.urlopen(done_nodes_url).read() # free host(s)
+if output != 'Done':
+	print "Error freeing host: %s" % host
+
+sys.exit(r)

misc/copr-build.py

@@ -0,0 +1,66 @@
+#!/usr/bin/python
+
+# README:
+# for initial setup, browse to: https://copr.fedoraproject.org/api/
+# and it will have a ~/.config/copr config that you can download.
+# happy hacking!
+
+import os
+import sys
+import copr
+import time
+
+COPR = 'mgmt'
+
+if len(sys.argv) != 2:
+	print("Usage: %s <srpm url>" % sys.argv[0])
+	sys.exit(1)
+
+url = sys.argv[1]
+
+client = copr.CoprClient.create_from_file_config(os.path.expanduser("~/.config/copr"))
+
+result = client.create_new_build(COPR, [url])
+if result.output != 'ok':
+	print(result.error)
+	sys.exit(1)
+print(result.message)
+
+# modified from: https://python-copr.readthedocs.org/en/latest/Examples.html#work-with-builds
+for bw in result.builds_list:
+	print("Build #{}: {}".format(bw.build_id, bw.handle.get_build_details().status))
+
+# cancel all created build
+#for bw in result.builds_list:
+#	bw.handle.cancel_build()
+
+# get build status for each chroot
+#for bw in result.builds_list:
+#	print("build: {}".format(bw.build_id))
+#	for ch, status in bw.handle.get_build_details().data["chroots"].items():
+#		print("\t chroot {}:\t {}".format(ch, status))
+
+# simple build progress:
+
+watched = set(result.builds_list)
+done = set()
+state = {}
+for bw in watched:	# store initial states
+	state[bw.build_id] = bw.handle.get_build_details().status
+
+while watched != done:
+	for bw in watched:
+		if bw in done:
+			continue
+		status = bw.handle.get_build_details().status
+		if status != state.get(bw.build_id):
+			print("Build #{}: {}".format(bw.build_id, status))
+			state[bw.build_id] = status	# update status
+
+		if status in ['skipped', 'failed', 'succeeded']:
+			done.add(bw)
+
+	if watched == done: break	# avoid long while sleep
+	else: time.sleep(10)
+
+print 'Done!'

misc/make-deps.sh

@@ -0,0 +1,39 @@
+#!/bin/bash
+# setup a simple go environment
+
+travis=0
+if env | grep -q '^TRAVIS=true$'; then
+	travis=1
+fi
+
+if [ $travis -eq 0 ]; then
+	YUM=`which yum 2>/dev/null`
+	APT=`which apt-get 2>/dev/null`
+	if [ -z "$YUM" -a -z "$APT" ]; then
+		echo "The package managers can't be found."
+		exit 1
+	fi
+	if [ ! -z "$YUM" ]; then
+		# some go dependencies are stored in mercurial
+		sudo $YUM install -y golang golang-googlecode-tools-stringer hg
+
+	fi
+	if [ ! -z "$APT" ]; then
+		sudo $APT install -y golang mercurial
+
+	fi
+fi
+
+# build etcd
+git clone --recursive https://github.com/coreos/etcd/ && cd etcd
+git checkout v2.2.4	# TODO: update to newer versions as needed
+[ -x build ] && ./build
+mkdir -p ~/bin/
+cp bin/etcd ~/bin/
+cd -
+rm -rf etcd	# clean up to avoid failing on upstream gofmt errors
+
+go get ./...	# get all the go dependencies
+[ -e "$GOBIN/mgmt" ] && rm -f "$GOBIN/mgmt"	# the `go get` version has no -X
+go get golang.org/x/tools/cmd/vet	# add in `go vet` for travis
+go get golang.org/x/tools/cmd/stringer	# for automatic stringer-ing

misc/make-path.sh

@@ -0,0 +1,48 @@
+#!/bin/bash
+# setup a few environment path values
+
+if ! env | grep -q '^GOPATH='; then
+	export GOPATH="$HOME/gopath/"
+	mkdir "$GOPATH"
+	if ! grep -q '^export GOPATH=' ~/.bashrc; then
+	        echo "export GOPATH=~/gopath/" >> ~/.bashrc
+	fi
+	echo "setting go path to: $GOPATH"
+fi
+
+echo "gopath is: $GOPATH"
+
+# some versions of golang apparently require this to run go get :(
+if ! env | grep -q '^GOBIN='; then
+	export GOBIN="${GOPATH}bin/"
+	mkdir "$GOBIN"
+	if ! grep -q '^export GOBIN=' ~/.bashrc; then
+	        echo 'export GOBIN="${GOPATH}bin/"' >> ~/.bashrc
+	fi
+	echo "setting go bin to: $GOBIN"
+fi
+
+echo "gobin is: $GOBIN"
+
+# add gobin to $PATH
+if ! env | grep '^PATH=' | grep -q "$GOBIN"; then
+	if ! grep -q '^export PATH="'"${GOBIN}"':${PATH}"' ~/.bashrc; then
+	        echo 'export PATH="'"${GOBIN}"':${PATH}"' >> ~/.bashrc
+	fi
+	export PATH="${GOBIN}:${PATH}"
+	echo "setting path to: $PATH"
+fi
+
+echo "path is: $PATH"
+
+# add ~/bin/ to $PATH
+if ! env | grep '^PATH=' | grep -q "$HOME/bin"; then
+	mkdir -p "${HOME}/bin"
+	if ! grep -q '^export PATH="'"${HOME}/bin"':${PATH}"' ~/.bashrc; then
+	        echo 'export PATH="'"${HOME}/bin"':${PATH}"' >> ~/.bashrc
+	fi
+	export PATH="${HOME}/bin:${PATH}"
+	echo "setting path to: $PATH"
+fi
+
+echo "path is: $PATH"

misc/mgmt.bashrc

@@ -0,0 +1,13 @@
+#!/bin/bash
+
+_cli_bash_autocomplete_mgmt() {
+	local cur prev opts base
+	COMPREPLY=()
+	cur="${COMP_WORDS[COMP_CWORD]}"
+	prev="${COMP_WORDS[COMP_CWORD-1]}"
+	opts=$( ${COMP_WORDS[@]:0:$COMP_CWORD} --generate-bash-completion )
+	COMPREPLY=( $(compgen -W "${opts}" -- ${cur}) )
+	return 0
+}
+
+complete -F _cli_bash_autocomplete_mgmt mgmt

misc/mgmt.conf.example

@@ -0,0 +1 @@
+# example mgmt configuration file, currently has not options at the moment!

misc_test.go

@@ -1,5 +1,5 @@
 // Mgmt
-// Copyright (C) 2013-2015+ James Shubin and the project contributors
+// Copyright (C) 2013-2016+ James Shubin and the project contributors
 // Written by James Shubin <james@shubin.ca> and the project contributors
 //
 // This program is free software: you can redistribute it and/or modify
@@ -18,6 +18,7 @@
 package main
 
 import (
+	"fmt"
 	"testing"
 )
 
@@ -31,9 +32,22 @@ func TestMiscT1(t *testing.T) {
 		t.Errorf("Result is incorrect.")
 	}
 
-	if Dirname("/") != "/" {
+	if Dirname("/") != "" { // TODO: should this equal "/" or "" ?
 		t.Errorf("Result is incorrect.")
 	}
+
+	if Basename("/foo/bar/baz") != "baz" {
+		t.Errorf("Result is incorrect.")
+	}
+
+	if Basename("/foo/bar/baz/") != "baz/" {
+		t.Errorf("Result is incorrect.")
+	}
+
+	if Basename("/") != "/" { // TODO: should this equal "" or "/" ?
+		t.Errorf("Result is incorrect.")
+	}
+
 }
 
 func TestMiscT2(t *testing.T) {
@@ -82,6 +96,41 @@ func TestMiscT3(t *testing.T) {
 
 func TestMiscT4(t *testing.T) {
 
+	if PathPrefixDelta("/foo/bar/baz", "/foo/ba") != -1 {
+		t.Errorf("Result should be -1.")
+	}
+
+	if PathPrefixDelta("/foo/bar/baz", "/foo/bar") != 1 {
+		t.Errorf("Result should be 1.")
+	}
+
+	if PathPrefixDelta("/foo/bar/baz", "/foo/bar/") != 1 {
+		t.Errorf("Result should be 1.")
+	}
+
+	if PathPrefixDelta("/foo/bar/baz/", "/foo/bar") != 1 {
+		t.Errorf("Result should be 1.")
+	}
+
+	if PathPrefixDelta("/foo/bar/baz/", "/foo/bar/") != 1 {
+		t.Errorf("Result should be 1.")
+	}
+
+	if PathPrefixDelta("/foo/bar/baz/", "/foo/bar/baz/dude") != -1 {
+		t.Errorf("Result should be -1.")
+	}
+
+	if PathPrefixDelta("/foo/bar/baz/a/b/c/", "/foo/bar/baz") != 3 {
+		t.Errorf("Result should be 3.")
+	}
+
+	if PathPrefixDelta("/foo/bar/baz/", "/foo/bar/baz") != 0 {
+		t.Errorf("Result should be 0.")
+	}
+}
+
+func TestMiscT5(t *testing.T) {
+
 	if PathIsDir("/foo/bar/baz/") != true {
 		t.Errorf("Result should be false.")
 	}
@@ -97,5 +146,55 @@ func TestMiscT4(t *testing.T) {
 	if PathIsDir("/") != true {
 		t.Errorf("Result should be true.")
 	}
-
+}
+
+func TestMiscT6(t *testing.T) {
+
+	type foo struct {
+		Name  string `yaml:"name"`
+		Type  string `yaml:"type"`
+		Value int    `yaml:"value"`
+	}
+
+	obj := foo{"dude", "sweet", 42}
+	output, ok := ObjToB64(obj)
+	if ok != true {
+		t.Errorf("First result should be true.")
+	}
+	var data foo
+	if B64ToObj(output, &data) != true {
+		t.Errorf("Second result should be true.")
+	}
+	// TODO: there is probably a better way to compare these two...
+	if fmt.Sprintf("%+v\n", obj) != fmt.Sprintf("%+v\n", data) {
+		t.Errorf("Strings should match.")
+	}
+}
+
+func TestMiscT7(t *testing.T) {
+
+	type Foo struct {
+		Name  string `yaml:"name"`
+		Type  string `yaml:"type"`
+		Value int    `yaml:"value"`
+	}
+
+	type bar struct {
+		Foo     `yaml:",inline"` // anonymous struct must be public!
+		Comment string           `yaml:"comment"`
+	}
+
+	obj := bar{Foo{"dude", "sweet", 42}, "hello world"}
+	output, ok := ObjToB64(obj)
+	if ok != true {
+		t.Errorf("First result should be true.")
+	}
+	var data bar
+	if B64ToObj(output, &data) != true {
+		t.Errorf("Second result should be true.")
+	}
+	// TODO: there is probably a better way to compare these two...
+	if fmt.Sprintf("%+v\n", obj) != fmt.Sprintf("%+v\n", data) {
+		t.Errorf("Strings should match.")
+	}
 }

omv.yaml

@@ -0,0 +1,39 @@
+---
+:domain: example.com
+:network: 192.168.123.0/24
+:image: fedora-23
+:cpus: ''
+:memory: ''
+:disks: 0
+:disksize: 40G
+:boxurlprefix: ''
+:sync: rsync
+:syncdir: mgmt/
+:syncsrc: "../"
+:folder: ".omv"
+:extern: []
+:cd: "-"
+:puppet: false
+:classes: []
+:shell:
+- cd /vagrant/mgmt/ && make deps
+:docker: false
+:kubernetes: false
+:ansible: []
+:playbook: []
+:ansible_extras: {}
+:cachier: false
+:vms: []
+:namespace: omv
+:count: 1
+:username: ''
+:password: ''
+:poolid: true
+:repos: []
+:update: false
+:reboot: false
+:unsafe: false
+:nested: false
+:tests: []
+:comment: ''
+:reallyrm: false

pgraph.go

@@ -1,5 +1,5 @@
 // Mgmt
-// Copyright (C) 2013-2015+ James Shubin and the project contributors
+// Copyright (C) 2013-2016+ James Shubin and the project contributors
 // Written by James Shubin <james@shubin.ca> and the project contributors
 //
 // This program is free software: you can redistribute it and/or modify
@@ -19,69 +19,106 @@
 package main
 
 import (
-	"code.google.com/p/go-uuid/uuid"
-	//"container/list" // doubly linked list
+	"errors"
 	"fmt"
+	"io/ioutil"
 	"log"
+	"os"
+	"os/exec"
+	"strconv"
 	"sync"
+	"syscall"
 	"time"
 )
 
+//go:generate stringer -type=graphState -output=graphstate_stringer.go
+type graphState int
+
+const (
+	graphNil graphState = iota
+	graphStarting
+	graphStarted
+	graphPausing
+	graphPaused
+)
+
 // The graph abstract data type (ADT) is defined as follows:
-// NOTE: the directed graph arrows point from left to right ( --> )
-// NOTE: the arrows point towards their dependencies (eg: arrows mean requires)
+// * the directed graph arrows point from left to right ( -> )
+// * the arrows point away from their dependencies (eg: arrows mean "before")
+// * IOW, you might see package -> file -> service (where package runs first)
+// * This is also the direction that the notify should happen in...
 type Graph struct {
-	uuid      string
 	Name      string
-	Adjacency map[*Vertex]map[*Vertex]*Edge
+	Adjacency map[*Vertex]map[*Vertex]*Edge // *Vertex -> *Vertex (edge)
+	state     graphState
+	mutex     sync.Mutex // used when modifying graph State variable
 	//Directed  bool
-	startcount int
 }
 
 type Vertex struct {
-	uuid      string
-	graph     *Graph // store a pointer to the graph it's on
-	Name      string
-	Type      string
-	Timestamp int64       // last updated timestamp ?
-	Events    chan string // FIXME: eventually a struct for the event?
-	Typedata  Type
-	data      map[string]string
+	graph *Graph            // store a pointer to the graph it's on
+	Type                    // anonymous field
+	data  map[string]string // XXX: currently unused i think, remove?
 }
 
 type Edge struct {
-	uuid string
 	Name string
 }
 
 func NewGraph(name string) *Graph {
 	return &Graph{
-		uuid:      uuid.New(),
 		Name:      name,
 		Adjacency: make(map[*Vertex]map[*Vertex]*Edge),
+		state:     graphNil,
 	}
 }
 
-func NewVertex(name, t string) *Vertex {
+func NewVertex(t Type) *Vertex {
 	return &Vertex{
-		uuid:      uuid.New(),
-		Name:      name,
-		Type:      t,
-		Timestamp: -1,
-		Events:    make(chan string, 1), // XXX: chan size?
-		data:      make(map[string]string),
+		Type: t,
+		data: make(map[string]string),
 	}
 }
 
 func NewEdge(name string) *Edge {
 	return &Edge{
-		uuid: uuid.New(),
 		Name: name,
 	}
 }
 
-// Graph() creates a new, empty graph.
-// addVertex(vert) adds an instance of Vertex to the graph.
+// returns the name of the graph
+func (g *Graph) GetName() string {
+	return g.Name
+}
+
+// set name of the graph
+func (g *Graph) SetName(name string) {
+	g.Name = name
+}
+
+func (g *Graph) GetState() graphState {
+	//g.mutex.Lock()
+	//defer g.mutex.Unlock()
+	return g.state
+}
+
+// set graph state and return previous state
+func (g *Graph) SetState(state graphState) graphState {
+	g.mutex.Lock()
+	defer g.mutex.Unlock()
+	prev := g.GetState()
+	g.state = state
+	return prev
+}
+
+// store a pointer in the type to it's parent vertex
+func (g *Graph) SetVertex() {
+	for v := range g.GetVerticesChan() {
+		v.Type.SetVertex(v)
+	}
+}
+
+// add a new vertex to the graph
 func (g *Graph) AddVertex(v *Vertex) {
 	if _, exists := g.Adjacency[v]; !exists {
 		g.Adjacency[v] = make(map[*Vertex]*Edge)
@@ -91,7 +128,14 @@ func (g *Graph) AddVertex(v *Vertex) {
 	}
 }
 
-// addEdge(fromVert, toVert) Adds a new, directed edge to the graph that connects two vertices.
+func (g *Graph) DeleteVertex(v *Vertex) {
+	delete(g.Adjacency, v)
+	for k := range g.Adjacency {
+		delete(g.Adjacency[k], v)
+	}
+}
+
+// adds a directed edge to the graph from v1 to v2
 func (g *Graph) AddEdge(v1, v2 *Vertex, e *Edge) {
 	// NOTE: this doesn't allow more than one edge between two vertexes...
 	// TODO: is this a problem?
@@ -100,35 +144,60 @@ func (g *Graph) AddEdge(v1, v2 *Vertex, e *Edge) {
 	g.Adjacency[v1][v2] = e
 }
 
-// addEdge(fromVert, toVert, weight) Adds a new, weighted, directed edge to the graph that connects two vertices.
-// getVertex(vertKey) finds the vertex in the graph named vertKey.
-func (g *Graph) GetVertex(uuid string) chan *Vertex {
+// XXX: does it make sense to return a channel here?
+// GetVertex finds the vertex in the graph with a particular search name
+func (g *Graph) GetVertex(name string) chan *Vertex {
 	ch := make(chan *Vertex, 1)
-	go func(uuid string) {
+	go func(name string) {
 		for k := range g.Adjacency {
-			v := *k
-			if v.uuid == uuid {
+			if k.GetName() == name {
 				ch <- k
 				break
 			}
 		}
 		close(ch)
-	}(uuid)
+	}(name)
 	return ch
 }
 
+func (g *Graph) GetVertexMatch(obj Type) *Vertex {
+	for k := range g.Adjacency {
+		if k.Compare(obj) { // XXX test
+			return k
+		}
+	}
+	return nil
+}
+
+func (g *Graph) HasVertex(v *Vertex) bool {
+	if _, exists := g.Adjacency[v]; exists {
+		return true
+	}
+	//for k := range g.Adjacency {
+	//	if k == v {
+	//		return true
+	//	}
+	//}
+	return false
+}
+
+// number of vertices in the graph
 func (g *Graph) NumVertices() int {
 	return len(g.Adjacency)
 }
 
+// number of edges in the graph
 func (g *Graph) NumEdges() int {
-	// XXX: not implemented
-	return -1
+	count := 0
+	for k := range g.Adjacency {
+		count += len(g.Adjacency[k])
+	}
+	return count
 }
 
 // get an array (slice) of all vertices in the graph
 func (g *Graph) GetVertices() []*Vertex {
-	vertices := make([]*Vertex, 0)
+	var vertices []*Vertex
 	for k := range g.Adjacency {
 		vertices = append(vertices, k)
 	}
@@ -138,7 +207,6 @@ func (g *Graph) GetVertices() []*Vertex {
 // returns a channel of all vertices in the graph
 func (g *Graph) GetVerticesChan() chan *Vertex {
 	ch := make(chan *Vertex)
-	// TODO: do you need to pass this through into the go routine?
 	go func(ch chan *Vertex) {
 		for k := range g.Adjacency {
 			ch <- k
@@ -153,7 +221,88 @@ func (g *Graph) String() string {
 	return fmt.Sprintf("Vertices(%d), Edges(%d)", g.NumVertices(), g.NumEdges())
 }
 
-//func (s []*Vertex) contains(element *Vertex) bool {
+// output the graph in graphviz format
+// https://en.wikipedia.org/wiki/DOT_%28graph_description_language%29
+func (g *Graph) Graphviz() (out string) {
+	//digraph g {
+	//	label="hello world";
+	//	node [shape=box];
+	//	A [label="A"];
+	//	B [label="B"];
+	//	C [label="C"];
+	//	D [label="D"];
+	//	E [label="E"];
+	//	A -> B [label=f];
+	//	B -> C [label=g];
+	//	D -> E [label=h];
+	//}
+	out += fmt.Sprintf("digraph %v {\n", g.GetName())
+	out += fmt.Sprintf("\tlabel=\"%v\";\n", g.GetName())
+	//out += "\tnode [shape=box];\n"
+	str := ""
+	for i := range g.Adjacency { // reverse paths
+		out += fmt.Sprintf("\t%v [label=\"%v[%v]\"];\n", i.GetName(), i.GetType(), i.GetName())
+		for j := range g.Adjacency[i] {
+			k := g.Adjacency[i][j]
+			// use str for clearer output ordering
+			str += fmt.Sprintf("\t%v -> %v [label=%v];\n", i.GetName(), j.GetName(), k.Name)
+		}
+	}
+	out += str
+	out += "}\n"
+	return
+}
+
+// write out the graphviz data and run the correct graphviz filter command
+func (g *Graph) ExecGraphviz(program, filename string) error {
+
+	switch program {
+	case "dot", "neato", "twopi", "circo", "fdp":
+	default:
+		return errors.New("Invalid graphviz program selected!")
+	}
+
+	if filename == "" {
+		return errors.New("No filename given!")
+	}
+
+	// run as a normal user if possible when run with sudo
+	uid, err1 := strconv.Atoi(os.Getenv("SUDO_UID"))
+	gid, err2 := strconv.Atoi(os.Getenv("SUDO_GID"))
+
+	err := ioutil.WriteFile(filename, []byte(g.Graphviz()), 0644)
+	if err != nil {
+		return errors.New("Error writing to filename!")
+	}
+
+	if err1 == nil && err2 == nil {
+		if err := os.Chown(filename, uid, gid); err != nil {
+			return errors.New("Error changing file owner!")
+		}
+	}
+
+	path, err := exec.LookPath(program)
+	if err != nil {
+		return errors.New("Graphviz is missing!")
+	}
+
+	out := fmt.Sprintf("%v.png", filename)
+	cmd := exec.Command(path, "-Tpng", fmt.Sprintf("-o%v", out), filename)
+
+	if err1 == nil && err2 == nil {
+		cmd.SysProcAttr = &syscall.SysProcAttr{}
+		cmd.SysProcAttr.Credential = &syscall.Credential{
+			Uid: uint32(uid),
+			Gid: uint32(gid),
+		}
+	}
+	_, err = cmd.Output()
+	if err != nil {
+		return errors.New("Error writing to image!")
+	}
+	return nil
+}
+
 // google/golang hackers apparently do not think contains should be a built-in!
 func Contains(s []*Vertex, element *Vertex) bool {
 	for _, v := range s {
@@ -164,20 +313,15 @@ func Contains(s []*Vertex, element *Vertex) bool {
 	return false
 }
 
-// return an array (slice) of all vertices that connect to vertex v
-func (g *Graph) GraphEdges(vertex *Vertex) []*Vertex {
+// return an array (slice) of all directed vertices to vertex v (??? -> v)
+// ostimestamp should use this
+func (g *Graph) IncomingGraphEdges(v *Vertex) []*Vertex {
 	// TODO: we might be able to implement this differently by reversing
 	// the Adjacency graph and then looping through it again...
-	s := make([]*Vertex, 0)                 // stack
-	for w, _ := range g.Adjacency[vertex] { // forward paths
-		//fmt.Printf("forward: %v -> %v\n", v.Name, w.Name)
-		s = append(s, w)
-	}
-
-	for k, x := range g.Adjacency { // reverse paths
-		for w, _ := range x {
-			if w == vertex {
-				//fmt.Printf("reverse: %v -> %v\n", v.Name, k.Name)
+	var s []*Vertex
+	for k := range g.Adjacency { // reverse paths
+		for w := range g.Adjacency[k] {
+			if w == v {
 				s = append(s, k)
 			}
 		}
@@ -185,32 +329,27 @@ func (g *Graph) GraphEdges(vertex *Vertex) []*Vertex {
 	return s
 }
 
-// return an array (slice) of all directed vertices to vertex v
-func (g *Graph) DirectedGraphEdges(vertex *Vertex) []*Vertex {
-	// TODO: we might be able to implement this differently by reversing
-	// the Adjacency graph and then looping through it again...
-	s := make([]*Vertex, 0)                 // stack
-	for w, _ := range g.Adjacency[vertex] { // forward paths
-		//fmt.Printf("forward: %v -> %v\n", v.Name, w.Name)
-		s = append(s, w)
+// return an array (slice) of all vertices that vertex v points to (v -> ???)
+// poke should use this
+func (g *Graph) OutgoingGraphEdges(v *Vertex) []*Vertex {
+	var s []*Vertex
+	for k := range g.Adjacency[v] { // forward paths
+		s = append(s, k)
 	}
 	return s
 }
 
-// get timestamp of a vertex
-func (v *Vertex) GetTimestamp() int64 {
-	return v.Timestamp
-}
-
-// update timestamp of a vertex
-func (v *Vertex) UpdateTimestamp() int64 {
-	v.Timestamp = time.Now().UnixNano() // update
-	return v.Timestamp
+// return an array (slice) of all vertices that connect to vertex v
+func (g *Graph) GraphEdges(v *Vertex) []*Vertex {
+	var s []*Vertex
+	s = append(s, g.IncomingGraphEdges(v)...)
+	s = append(s, g.OutgoingGraphEdges(v)...)
+	return s
 }
 
 func (g *Graph) DFS(start *Vertex) []*Vertex {
-	d := make([]*Vertex, 0) // discovered
-	s := make([]*Vertex, 0) // stack
+	var d []*Vertex // discovered
+	var s []*Vertex // stack
 	if _, exists := g.Adjacency[start]; !exists {
 		return nil // TODO: error
 	}
@@ -238,7 +377,7 @@ func (g *Graph) FilterGraph(name string, vertices []*Vertex) *Graph {
 	for k1, x := range g.Adjacency {
 		for k2, e := range x {
 
-			//fmt.Printf("Filter: %v -> %v # %v\n", k1.Name, k2.Name, e.Name)
+			//log.Printf("Filter: %v -> %v # %v", k1.Name, k2.Name, e.Name)
 			if Contains(vertices, k1) || Contains(vertices, k2) {
 				newgraph.AddEdge(k1, k2, e)
 			}
@@ -254,7 +393,7 @@ func (g *Graph) GetDisconnectedGraphs() chan *Graph {
 	ch := make(chan *Graph)
 	go func() {
 		var start *Vertex
-		d := make([]*Vertex, 0) // discovered
+		var d []*Vertex // discovered
 		c := g.NumVertices()
 		for len(d) < c {
 
@@ -279,12 +418,91 @@ func (g *Graph) GetDisconnectedGraphs() chan *Graph {
 			// if we've found all the elements, then we're done
 			// otherwise loop through to continue...
 		}
-
 		close(ch)
 	}()
 	return ch
 }
 
+// return the indegree for the graph, IOW the count of vertices that point to me
+// NOTE: this returns the values for all vertices in one big lookup table
+func (g *Graph) InDegree() map[*Vertex]int {
+	result := make(map[*Vertex]int)
+	for k := range g.Adjacency {
+		result[k] = 0 // initialize
+	}
+
+	for k := range g.Adjacency {
+		for z := range g.Adjacency[k] {
+			result[z]++
+		}
+	}
+	return result
+}
+
+// return the outdegree for the graph, IOW the count of vertices that point away
+// NOTE: this returns the values for all vertices in one big lookup table
+func (g *Graph) OutDegree() map[*Vertex]int {
+	result := make(map[*Vertex]int)
+
+	for k := range g.Adjacency {
+		result[k] = 0 // initialize
+		for _ = range g.Adjacency[k] {
+			result[k]++
+		}
+	}
+	return result
+}
+
+// returns a topological sort for the graph
+// based on descriptions and code from wikipedia and rosetta code
+// TODO: add memoization, and cache invalidation to speed this up :)
+func (g *Graph) TopologicalSort() (result []*Vertex, ok bool) { // kahn's algorithm
+
+	var L []*Vertex                    // empty list that will contain the sorted elements
+	var S []*Vertex                    // set of all nodes with no incoming edges
+	remaining := make(map[*Vertex]int) // amount of edges remaining
+
+	for v, d := range g.InDegree() {
+		if d == 0 {
+			// accumulate set of all nodes with no incoming edges
+			S = append(S, v)
+		} else {
+			// initialize remaining edge count from indegree
+			remaining[v] = d
+		}
+	}
+
+	for len(S) > 0 {
+		last := len(S) - 1 // remove a node v from S
+		v := S[last]
+		S = S[:last]
+		L = append(L, v) // add v to tail of L
+		for n := range g.Adjacency[v] {
+			// for each node n remaining in the graph, consume from
+			// remaining, so for remaining[n] > 0
+			if remaining[n] > 0 {
+				remaining[n]--         // remove edge from the graph
+				if remaining[n] == 0 { // if n has no other incoming edges
+					S = append(S, n) // insert n into S
+				}
+			}
+		}
+	}
+
+	// if graph has edges, eg if any value in rem is > 0
+	for c, in := range remaining {
+		if in > 0 {
+			for n := range g.Adjacency[c] {
+				if remaining[n] > 0 {
+					return nil, false // not a dag!
+				}
+			}
+		}
+	}
+
+	return L, true
+}
+
 func (v *Vertex) Value(key string) (string, bool) {
 	if value, exists := v.data[key]; exists {
 		return value, true
@@ -324,91 +542,93 @@ func HeisenbergCount(ch chan *Vertex) int {
 	return c
 }
 
-func (v *Vertex) Associate(t Type) {
-	v.Typedata = t
-}
+// main kick to start the graph
+func (g *Graph) Start(wg *sync.WaitGroup, first bool) { // start or continue
+	t, _ := g.TopologicalSort()
+	// TODO: only calculate indegree if `first` is true to save resources
+	indegree := g.InDegree() // compute all of the indegree's
+	for _, v := range Reverse(t) {
 
-func (v *Vertex) OKTimestamp() bool {
-	g := v.GetGraph()
-	for _, n := range g.DirectedGraphEdges(v) {
-		if v.GetTimestamp() > n.GetTimestamp() {
-			return false
+		if !v.Type.IsWatching() { // if Watch() is not running...
+			wg.Add(1)
+			// must pass in value to avoid races...
+			// see: https://ttboj.wordpress.com/2015/07/27/golang-parallelism-issues-causing-too-many-open-files-error/
+			go func(vv *Vertex) {
+				defer wg.Done()
+				vv.Type.Watch()
+				log.Printf("%v[%v]: Exited", vv.GetType(), vv.GetName())
+			}(v)
+		}
+
+		// selective poke: here we reduce the number of initial pokes
+		// to the minimum required to activate every vertex in the
+		// graph, either by direct action, or by getting poked by a
+		// vertex that was previously activated. if we poke each vertex
+		// that has no incoming edges, then we can be sure to reach the
+		// whole graph. Please note: this may mask certain optimization
+		// failures, such as any poke limiting code in Poke() or
+		// BackPoke(). You might want to disable this selective start
+		// when experimenting with and testing those elements.
+		// if we are unpausing (since it's not the first run of this
+		// function) we need to poke to *unpause* every graph vertex,
+		// and not just selectively the subset with no indegree.
+		if (!first) || indegree[v] == 0 {
+			// ensure state is started before continuing on to next vertex
+			for !v.Type.SendEvent(eventStart, true, false) {
+				if DEBUG {
+					// if SendEvent fails, we aren't up yet
+					log.Printf("%v[%v]: Retrying SendEvent(Start)", v.GetType(), v.GetName())
+					// sleep here briefly or otherwise cause
+					// a different goroutine to be scheduled
+					time.Sleep(1 * time.Millisecond)
+				}
+			}
 		}
 	}
-
-	return true
 }
 
-// poke the XXX children?
-func (v *Vertex) Poke() {
-	g := v.GetGraph()
-
-	for _, n := range g.DirectedGraphEdges(v) { // XXX: do we want the reverse order?
-		// poke!
-		n.Events <- fmt.Sprintf("poke(%v)", v.Name)
+func (g *Graph) Pause() {
+	t, _ := g.TopologicalSort()
+	for _, v := range t { // squeeze out the events...
+		v.Type.SendEvent(eventPause, true, false)
 	}
 }
 
 func (g *Graph) Exit() {
-	// tell all the vertices to exit...
-	for v := range g.GetVerticesChan() {
-		v.Exit()
+	t, _ := g.TopologicalSort()
+	for _, v := range t { // squeeze out the events...
+		// turn off the taps...
+		// XXX: do this by sending an exit signal, and then returning
+		// when we hit the 'default' in the select statement!
+		// XXX: we can do this to quiesce, but it's not necessary now
+
+		v.Type.SendEvent(eventExit, true, false)
 	}
 }
 
-func (v *Vertex) Exit() {
-	v.Events <- "exit"
+func (g *Graph) SetConvergedCallback(ctimeout int, converged chan bool) {
+	for v := range g.GetVerticesChan() {
+		v.Type.SetConvegedCallback(ctimeout, converged)
+	}
 }
 
-// main loop for each vertex
-// warning: this logic might be subtle and tricky.
-// be careful as it might not even be correct now!
-func (v *Vertex) Start() {
-	log.Printf("Main->Vertex[%v]->Start()\n", v.Name)
-
-	//g := v.GetGraph()
-	var t = v.Typedata
-
-	// this whole wg2 wait group is only necessary if we need to wait for
-	// the go routine to exit...
-	var wg2 sync.WaitGroup
-
-	wg2.Add(1)
-	go func(v *Vertex, t Type) {
-		defer wg2.Done()
-		//fmt.Printf("About to watch [%v].\n", v.Name)
-		t.Watch(v)
-	}(v, t)
-
-	var ok bool
-	//XXX make sure dependencies run and become more current first...
-	for {
-		select {
-		case event := <-v.Events:
-
-			log.Printf("Event[%v]: %v\n", v.Name, event)
-
-			if event == "exit" {
-				t.Exit()   // type exit
-				wg2.Wait() // wait for worker to exit
-				return
-			}
-
-			ok = true
-			if v.OKTimestamp() {
-				if !t.StateOK() { // TODO: can we rename this to something better?
-					// throw an error if apply fails...
-					// if this fails, don't UpdateTimestamp()
-					if !t.Apply() { // check for error
-						ok = false
-					}
-				}
-
-				if ok {
-					v.UpdateTimestamp() // this was touched...
-					v.Poke()            // XXX
-				}
-			}
+// in array function to test *vertices in a slice of *vertices
+func HasVertex(v *Vertex, haystack []*Vertex) bool {
+	for _, r := range haystack {
+		if v == r {
+			return true
 		}
 	}
+	return false
+}
+
+// reverse a list of vertices
+func Reverse(vs []*Vertex) []*Vertex {
+	//var out []*Vertex       // XXX: golint suggests, but it fails testing
+	out := make([]*Vertex, 0) // empty list
+	l := len(vs)
+	for i := range vs {
+		out = append(out, vs[l-i-1])
+	}
+	return out
 }

pgraph_test.go

@@ -1,5 +1,5 @@
 // Mgmt
-// Copyright (C) 2013-2015+ James Shubin and the project contributors
+// Copyright (C) 2013-2016+ James Shubin and the project contributors
 // Written by James Shubin <james@shubin.ca> and the project contributors
 //
 // This program is free software: you can redistribute it and/or modify
@@ -20,6 +20,7 @@
 package main
 
 import (
+	"reflect"
 	"testing"
 )
 
@@ -31,25 +32,33 @@ func TestPgraphT1(t *testing.T) {
 		t.Errorf("Should have 0 vertices instead of: %d.", i)
 	}
 
-	v1 := NewVertex("v1", "type")
-	v2 := NewVertex("v2", "type")
+	if i := G.NumEdges(); i != 0 {
+		t.Errorf("Should have 0 edges instead of: %d.", i)
+	}
+
+	v1 := NewVertex(NewNoopType("v1"))
+	v2 := NewVertex(NewNoopType("v2"))
 	e1 := NewEdge("e1")
 	G.AddEdge(v1, v2, e1)
 
 	if i := G.NumVertices(); i != 2 {
 		t.Errorf("Should have 2 vertices instead of: %d.", i)
 	}
+
+	if i := G.NumEdges(); i != 1 {
+		t.Errorf("Should have 1 edges instead of: %d.", i)
+	}
 }
 
 func TestPgraphT2(t *testing.T) {
 
 	G := NewGraph("g2")
-	v1 := NewVertex("v1", "type")
-	v2 := NewVertex("v2", "type")
-	v3 := NewVertex("v3", "type")
-	v4 := NewVertex("v4", "type")
-	v5 := NewVertex("v5", "type")
-	v6 := NewVertex("v6", "type")
+	v1 := NewVertex(NewNoopType("v1"))
+	v2 := NewVertex(NewNoopType("v2"))
+	v3 := NewVertex(NewNoopType("v3"))
+	v4 := NewVertex(NewNoopType("v4"))
+	v5 := NewVertex(NewNoopType("v5"))
+	v6 := NewVertex(NewNoopType("v6"))
 	e1 := NewEdge("e1")
 	e2 := NewEdge("e2")
 	e3 := NewEdge("e3")
@@ -71,12 +80,12 @@ func TestPgraphT2(t *testing.T) {
 func TestPgraphT3(t *testing.T) {
 
 	G := NewGraph("g3")
-	v1 := NewVertex("v1", "type")
-	v2 := NewVertex("v2", "type")
-	v3 := NewVertex("v3", "type")
-	v4 := NewVertex("v4", "type")
-	v5 := NewVertex("v5", "type")
-	v6 := NewVertex("v6", "type")
+	v1 := NewVertex(NewNoopType("v1"))
+	v2 := NewVertex(NewNoopType("v2"))
+	v3 := NewVertex(NewNoopType("v3"))
+	v4 := NewVertex(NewNoopType("v4"))
+	v5 := NewVertex(NewNoopType("v5"))
+	v6 := NewVertex(NewNoopType("v6"))
 	e1 := NewEdge("e1")
 	e2 := NewEdge("e2")
 	e3 := NewEdge("e3")
@@ -95,7 +104,7 @@ func TestPgraphT3(t *testing.T) {
 		t.Errorf("Should have 3 vertices instead of: %d.", i)
 		t.Errorf("Found: %v", out1)
 		for _, v := range out1 {
-			t.Errorf("Value: %v", v.Name)
+			t.Errorf("Value: %v", v.GetName())
 		}
 	}
 
@@ -104,7 +113,7 @@ func TestPgraphT3(t *testing.T) {
 		t.Errorf("Should have 3 vertices instead of: %d.", i)
 		t.Errorf("Found: %v", out1)
 		for _, v := range out1 {
-			t.Errorf("Value: %v", v.Name)
+			t.Errorf("Value: %v", v.GetName())
 		}
 	}
 }
@@ -112,9 +121,9 @@ func TestPgraphT3(t *testing.T) {
 func TestPgraphT4(t *testing.T) {
 
 	G := NewGraph("g4")
-	v1 := NewVertex("v1", "type")
-	v2 := NewVertex("v2", "type")
-	v3 := NewVertex("v3", "type")
+	v1 := NewVertex(NewNoopType("v1"))
+	v2 := NewVertex(NewNoopType("v2"))
+	v3 := NewVertex(NewNoopType("v3"))
 	e1 := NewEdge("e1")
 	e2 := NewEdge("e2")
 	e3 := NewEdge("e3")
@@ -127,19 +136,19 @@ func TestPgraphT4(t *testing.T) {
 		t.Errorf("Should have 3 vertices instead of: %d.", i)
 		t.Errorf("Found: %v", out)
 		for _, v := range out {
-			t.Errorf("Value: %v", v.Name)
+			t.Errorf("Value: %v", v.GetName())
 		}
 	}
 }
 
 func TestPgraphT5(t *testing.T) {
 	G := NewGraph("g5")
-	v1 := NewVertex("v1", "type")
-	v2 := NewVertex("v2", "type")
-	v3 := NewVertex("v3", "type")
-	v4 := NewVertex("v4", "type")
-	v5 := NewVertex("v5", "type")
-	v6 := NewVertex("v6", "type")
+	v1 := NewVertex(NewNoopType("v1"))
+	v2 := NewVertex(NewNoopType("v2"))
+	v3 := NewVertex(NewNoopType("v3"))
+	v4 := NewVertex(NewNoopType("v4"))
+	v5 := NewVertex(NewNoopType("v5"))
+	v6 := NewVertex(NewNoopType("v6"))
 	e1 := NewEdge("e1")
 	e2 := NewEdge("e2")
 	e3 := NewEdge("e3")
@@ -159,17 +168,16 @@ func TestPgraphT5(t *testing.T) {
 	if i := out.NumVertices(); i != 3 {
 		t.Errorf("Should have 3 vertices instead of: %d.", i)
 	}
-
 }
 
 func TestPgraphT6(t *testing.T) {
 	G := NewGraph("g6")
-	v1 := NewVertex("v1", "type")
-	v2 := NewVertex("v2", "type")
-	v3 := NewVertex("v3", "type")
-	v4 := NewVertex("v4", "type")
-	v5 := NewVertex("v5", "type")
-	v6 := NewVertex("v6", "type")
+	v1 := NewVertex(NewNoopType("v1"))
+	v2 := NewVertex(NewNoopType("v2"))
+	v3 := NewVertex(NewNoopType("v3"))
+	v4 := NewVertex(NewNoopType("v4"))
+	v5 := NewVertex(NewNoopType("v5"))
+	v6 := NewVertex(NewNoopType("v6"))
 	e1 := NewEdge("e1")
 	e2 := NewEdge("e2")
 	e3 := NewEdge("e3")
@@ -197,5 +205,204 @@ func TestPgraphT6(t *testing.T) {
 	if i := HeisenbergGraphCount(graphs); i != 2 {
 		t.Errorf("Should have 2 graphs instead of: %d.", i)
 	}
+}
+
+func TestPgraphT7(t *testing.T) {
+
+	G := NewGraph("g7")
+	v1 := NewVertex(NewNoopType("v1"))
+	v2 := NewVertex(NewNoopType("v2"))
+	v3 := NewVertex(NewNoopType("v3"))
+	e1 := NewEdge("e1")
+	e2 := NewEdge("e2")
+	e3 := NewEdge("e3")
+	G.AddEdge(v1, v2, e1)
+	G.AddEdge(v2, v3, e2)
+	G.AddEdge(v3, v1, e3)
+
+	if i := G.NumVertices(); i != 3 {
+		t.Errorf("Should have 3 vertices instead of: %d.", i)
+	}
+
+	G.DeleteVertex(v2)
+
+	if i := G.NumVertices(); i != 2 {
+		t.Errorf("Should have 2 vertices instead of: %d.", i)
+	}
+
+	G.DeleteVertex(v1)
+
+	if i := G.NumVertices(); i != 1 {
+		t.Errorf("Should have 1 vertices instead of: %d.", i)
+	}
+
+	G.DeleteVertex(v3)
+
+	if i := G.NumVertices(); i != 0 {
+		t.Errorf("Should have 0 vertices instead of: %d.", i)
+	}
+
+	G.DeleteVertex(v2) // duplicate deletes don't error...
+
+	if i := G.NumVertices(); i != 0 {
+		t.Errorf("Should have 0 vertices instead of: %d.", i)
+	}
+}
+
+func TestPgraphT8(t *testing.T) {
+
+	v1 := NewVertex(NewNoopType("v1"))
+	v2 := NewVertex(NewNoopType("v2"))
+	v3 := NewVertex(NewNoopType("v3"))
+	if HasVertex(v1, []*Vertex{v1, v2, v3}) != true {
+		t.Errorf("Should be true instead of false.")
+	}
+
+	v4 := NewVertex(NewNoopType("v4"))
+	v5 := NewVertex(NewNoopType("v5"))
+	v6 := NewVertex(NewNoopType("v6"))
+	if HasVertex(v4, []*Vertex{v5, v6}) != false {
+		t.Errorf("Should be false instead of true.")
+	}
+
+	v7 := NewVertex(NewNoopType("v7"))
+	v8 := NewVertex(NewNoopType("v8"))
+	v9 := NewVertex(NewNoopType("v9"))
+	if HasVertex(v8, []*Vertex{v7, v8, v9}) != true {
+		t.Errorf("Should be true instead of false.")
+	}
+
+	v1b := NewVertex(NewNoopType("v1")) // same value, different objects
+	if HasVertex(v1b, []*Vertex{v1, v2, v3}) != false {
+		t.Errorf("Should be false instead of true.")
+	}
+}
+
+func TestPgraphT9(t *testing.T) {
+
+	G := NewGraph("g9")
+	v1 := NewVertex(NewNoopType("v1"))
+	v2 := NewVertex(NewNoopType("v2"))
+	v3 := NewVertex(NewNoopType("v3"))
+	v4 := NewVertex(NewNoopType("v4"))
+	v5 := NewVertex(NewNoopType("v5"))
+	v6 := NewVertex(NewNoopType("v6"))
+	e1 := NewEdge("e1")
+	e2 := NewEdge("e2")
+	e3 := NewEdge("e3")
+	e4 := NewEdge("e4")
+	e5 := NewEdge("e5")
+	e6 := NewEdge("e6")
+	G.AddEdge(v1, v2, e1)
+	G.AddEdge(v1, v3, e2)
+	G.AddEdge(v2, v4, e3)
+	G.AddEdge(v3, v4, e4)
+
+	G.AddEdge(v4, v5, e5)
+	G.AddEdge(v5, v6, e6)
+
+	indegree := G.InDegree() // map[*Vertex]int
+	if i := indegree[v1]; i != 0 {
+		t.Errorf("Indegree of v1 should be 0 instead of: %d.", i)
+	}
+	if i := indegree[v2]; i != 1 {
+		t.Errorf("Indegree of v2 should be 1 instead of: %d.", i)
+	}
+	if i := indegree[v3]; i != 1 {
+		t.Errorf("Indegree of v3 should be 1 instead of: %d.", i)
+	}
+	if i := indegree[v4]; i != 2 {
+		t.Errorf("Indegree of v4 should be 2 instead of: %d.", i)
+	}
+	if i := indegree[v5]; i != 1 {
+		t.Errorf("Indegree of v5 should be 1 instead of: %d.", i)
+	}
+	if i := indegree[v6]; i != 1 {
+		t.Errorf("Indegree of v6 should be 1 instead of: %d.", i)
+	}
+
+	outdegree := G.OutDegree() // map[*Vertex]int
+	if i := outdegree[v1]; i != 2 {
+		t.Errorf("Outdegree of v1 should be 2 instead of: %d.", i)
+	}
+	if i := outdegree[v2]; i != 1 {
+		t.Errorf("Outdegree of v2 should be 1 instead of: %d.", i)
+	}
+	if i := outdegree[v3]; i != 1 {
+		t.Errorf("Outdegree of v3 should be 1 instead of: %d.", i)
+	}
+	if i := outdegree[v4]; i != 1 {
+		t.Errorf("Outdegree of v4 should be 1 instead of: %d.", i)
+	}
+	if i := outdegree[v5]; i != 1 {
+		t.Errorf("Outdegree of v5 should be 1 instead of: %d.", i)
+	}
+	if i := outdegree[v6]; i != 0 {
+		t.Errorf("Outdegree of v6 should be 0 instead of: %d.", i)
+	}
+
+	s, ok := G.TopologicalSort()
+	// either possibility is a valid toposort
+	match := reflect.DeepEqual(s, []*Vertex{v1, v2, v3, v4, v5, v6}) || reflect.DeepEqual(s, []*Vertex{v1, v3, v2, v4, v5, v6})
+	if !ok || !match {
+		t.Errorf("Topological sort failed, status: %v.", ok)
+		str := "Found:"
+		for _, v := range s {
+			str += " " + v.Type.GetName()
+		}
+		t.Errorf(str)
+	}
+}
+
+func TestPgraphT10(t *testing.T) {
+
+	G := NewGraph("g10")
+	v1 := NewVertex(NewNoopType("v1"))
+	v2 := NewVertex(NewNoopType("v2"))
+	v3 := NewVertex(NewNoopType("v3"))
+	v4 := NewVertex(NewNoopType("v4"))
+	v5 := NewVertex(NewNoopType("v5"))
+	v6 := NewVertex(NewNoopType("v6"))
+	e1 := NewEdge("e1")
+	e2 := NewEdge("e2")
+	e3 := NewEdge("e3")
+	e4 := NewEdge("e4")
+	e5 := NewEdge("e5")
+	e6 := NewEdge("e6")
+	G.AddEdge(v1, v2, e1)
+	G.AddEdge(v2, v3, e2)
+	G.AddEdge(v3, v4, e3)
+	G.AddEdge(v4, v5, e4)
+	G.AddEdge(v5, v6, e5)
+	G.AddEdge(v4, v2, e6) // cycle
+
+	if _, ok := G.TopologicalSort(); ok {
+		t.Errorf("Topological sort passed, but graph is cyclic.")
+	}
+}
+
+func TestPgraphT11(t *testing.T) {
+	v1 := NewVertex(NewNoopType("v1"))
+	v2 := NewVertex(NewNoopType("v2"))
+	v3 := NewVertex(NewNoopType("v3"))
+	v4 := NewVertex(NewNoopType("v4"))
+	v5 := NewVertex(NewNoopType("v5"))
+	v6 := NewVertex(NewNoopType("v6"))
+
+	if rev := Reverse([]*Vertex{}); !reflect.DeepEqual(rev, []*Vertex{}) {
+		t.Errorf("Reverse of vertex slice failed.")
+	}
+
+	if rev := Reverse([]*Vertex{v1}); !reflect.DeepEqual(rev, []*Vertex{v1}) {
+		t.Errorf("Reverse of vertex slice failed.")
+	}
+
+	if rev := Reverse([]*Vertex{v1, v2, v3, v4, v5, v6}); !reflect.DeepEqual(rev, []*Vertex{v6, v5, v4, v3, v2, v1}) {
+		t.Errorf("Reverse of vertex slice failed.")
+	}
+
+	if rev := Reverse([]*Vertex{v6, v5, v4, v3, v2, v1}); !reflect.DeepEqual(rev, []*Vertex{v1, v2, v3, v4, v5, v6}) {
+		t.Errorf("Reverse of vertex slice failed.")
+	}
 
 }

service.go

@@ -1,5 +1,5 @@
 // Mgmt
-// Copyright (C) 2013-2015+ James Shubin and the project contributors
+// Copyright (C) 2013-2016+ James Shubin and the project contributors
 // Written by James Shubin <james@shubin.ca> and the project contributors
 //
 // This program is free software: you can redistribute it and/or modify
@@ -20,7 +20,6 @@
 package main
 
 import (
-	"code.google.com/p/go-uuid/uuid"
 	"fmt"
 	systemd "github.com/coreos/go-systemd/dbus" // change namespace
 	"github.com/coreos/go-systemd/util"
@@ -29,29 +28,37 @@ import (
 )
 
 type ServiceType struct {
-	uuid    string
-	Type    string      // always "service"
-	Name    string      // name variable
-	Events  chan string // FIXME: eventually a struct for the event?
-	State   string      // state: running, stopped
-	Startup string      // enabled, disabled, undefined
+	BaseType `yaml:",inline"`
+	State    string `yaml:"state"`   // state: running, stopped
+	Startup  string `yaml:"startup"` // enabled, disabled, undefined
 }
 
 func NewServiceType(name, state, startup string) *ServiceType {
 	return &ServiceType{
-		uuid:    uuid.New(),
-		Type:    "service",
-		Name:    name,
-		Events:  make(chan string, 1), // XXX: chan size?
+		BaseType: BaseType{
+			Name:   name,
+			events: make(chan Event),
+			vertex: nil,
+		},
 		State:   state,
 		Startup: startup,
 	}
 }
 
-// Service watcher
-func (obj ServiceType) Watch(v *Vertex) {
-	// obj.Name: service name
+func (obj *ServiceType) GetType() string {
+	return "Service"
+}
 
+// Service watcher
+func (obj *ServiceType) Watch() {
+	if obj.IsWatching() {
+		return
+	}
+	obj.SetWatching(true)
+	defer obj.SetWatching(false)
+
+	// obj.Name: service name
+	//vertex := obj.GetVertex()         // stored with SetVertex
 	if !util.IsRunningSystemd() {
 		log.Fatal("Systemd is not running.")
 	}
@@ -64,7 +71,7 @@ func (obj ServiceType) Watch(v *Vertex) {
 
 	bus, err := dbus.SystemBus()
 	if err != nil {
-		log.Fatal("Failed to connect to bus: %v\n", err)
+		log.Fatal("Failed to connect to bus: ", err)
 	}
 
 	// XXX: will this detect new units?
@@ -75,9 +82,11 @@ func (obj ServiceType) Watch(v *Vertex) {
 
 	var service = fmt.Sprintf("%v.service", obj.Name) // systemd name
 	var send = false                                  // send event?
-	var invalid = false                               // does the service exist or not?
-	var previous bool                                 // previous invalid value
-	set := conn.NewSubscriptionSet()                  // no error should be returned
+	var exit = false
+	var dirty = false
+	var invalid = false              // does the service exist or not?
+	var previous bool                // previous invalid value
+	set := conn.NewSubscriptionSet() // no error should be returned
 	subChannel, subErrors := set.Subscribe()
 	var activeSet = false
 
@@ -91,40 +100,50 @@ func (obj ServiceType) Watch(v *Vertex) {
 		// firstly, does service even exist or not?
 		loadstate, err := conn.GetUnitProperty(service, "LoadState")
 		if err != nil {
-			log.Printf("Failed to get property: %v\n", err)
+			log.Printf("Failed to get property: %v", err)
 			invalid = true
 		}
 
 		if !invalid {
 			var notFound = (loadstate.Value == dbus.MakeVariant("not-found"))
 			if notFound { // XXX: in the loop we'll handle changes better...
-				log.Printf("Failed to find service: %v\n", service)
+				log.Printf("Failed to find service: %v", service)
 				invalid = true // XXX ?
 			}
 		}
 
 		if previous != invalid { // if invalid changed, send signal
 			send = true
+			dirty = true
 		}
 
 		if invalid {
-			log.Printf("Waiting for: %v\n", service) // waiting for service to appear...
+			log.Printf("Waiting for: %v", service) // waiting for service to appear...
 			if activeSet {
 				activeSet = false
 				set.Remove(service) // no return value should ever occur
 			}
 
+			obj.SetState(typeWatching) // reset
 			select {
 			case _ = <-buschan: // XXX wait for new units event to unstick
+				obj.SetConvergedState(typeConvergedNil)
 				// loop so that we can see the changed invalid signal
-				log.Printf("Service[%v]->DaemonReload()\n", service)
+				log.Printf("Service[%v]->DaemonReload()", service)
 
-			case exit := <-obj.Events:
-				if exit == "exit" {
-					return
-				} else {
-					log.Fatal("Unknown event: %v\n", exit)
+			case event := <-obj.events:
+				obj.SetConvergedState(typeConvergedNil)
+				if exit, send = obj.ReadEvent(&event); exit {
+					return // exit
 				}
+				if event.GetActivity() {
+					dirty = true
+				}
+
+			case _ = <-TimeAfterOrBlock(obj.ctimeout):
+				obj.SetConvergedState(typeConvergedTimeout)
+				obj.converged <- true
+				continue
 			}
 		} else {
 			if !activeSet {
@@ -132,55 +151,62 @@ func (obj ServiceType) Watch(v *Vertex) {
 				set.Add(service) // no return value should ever occur
 			}
 
-			log.Printf("Watching: %v\n", service) // attempting to watch...
+			log.Printf("Watching: %v", service) // attempting to watch...
+			obj.SetState(typeWatching)          // reset
 			select {
 			case event := <-subChannel:
 
-				log.Printf("Service event: %+v\n", event)
+				log.Printf("Service event: %+v", event)
 				// NOTE: the value returned is a map for some reason...
 				if event[service] != nil {
 					// event[service].ActiveState is not nil
 					if event[service].ActiveState == "active" {
-						log.Printf("Service[%v]->Started()\n", service)
+						log.Printf("Service[%v]->Started()", service)
 					} else if event[service].ActiveState == "inactive" {
-						log.Printf("Service[%v]->Stopped!()\n", service)
+						log.Printf("Service[%v]->Stopped!()", service)
 					} else {
 						log.Fatal("Unknown service state: ", event[service].ActiveState)
 					}
 				} else {
 					// service stopped (and ActiveState is nil...)
-					log.Printf("Service[%v]->Stopped\n", service)
+					log.Printf("Service[%v]->Stopped", service)
 				}
 				send = true
+				dirty = true
 
 			case err := <-subErrors:
+				obj.SetConvergedState(typeConvergedNil) // XXX ?
 				log.Println("error:", err)
 				log.Fatal(err)
-				v.Events <- fmt.Sprintf("service: %v", "error")
+				//vertex.events <- fmt.Sprintf("service: %v", "error") // XXX: how should we handle errors?
 
-			case exit := <-obj.Events:
-				if exit == "exit" {
-					return
-				} else {
-					log.Fatal("Unknown event: %v\n", exit)
+			case event := <-obj.events:
+				obj.SetConvergedState(typeConvergedNil)
+				if exit, send = obj.ReadEvent(&event); exit {
+					return // exit
+				}
+				if event.GetActivity() {
+					dirty = true
 				}
 			}
 		}
 
 		if send {
 			send = false
-			//log.Println("Sending event!")
-			v.Events <- fmt.Sprintf("service(%v): %v", obj.Name, "event!") // FIXME: use struct
+			if dirty {
+				dirty = false
+				obj.isStateOK = false // something made state dirty
+			}
+			Process(obj) // XXX: rename this function
 		}
+
 	}
 }
 
-func (obj ServiceType) Exit() bool {
-	obj.Events <- "exit"
-	return true
-}
-
-func (obj ServiceType) StateOK() bool {
+func (obj *ServiceType) StateOK() bool {
+	if obj.isStateOK { // cache the state
+		return true
+	}
 
 	if !util.IsRunningSystemd() {
 		log.Fatal("Systemd is not running.")
@@ -196,14 +222,14 @@ func (obj ServiceType) StateOK() bool {
 
 	loadstate, err := conn.GetUnitProperty(service, "LoadState")
 	if err != nil {
-		log.Printf("Failed to get load state: %v\n", err)
+		log.Printf("Failed to get load state: %v", err)
 		return false
 	}
 
 	// NOTE: we have to compare variants with other variants, they are really strings...
 	var notFound = (loadstate.Value == dbus.MakeVariant("not-found"))
 	if notFound {
-		log.Printf("Failed to find service: %v\n", service)
+		log.Printf("Failed to find service: %v", service)
 		return false
 	}
 
@@ -232,8 +258,8 @@ func (obj ServiceType) StateOK() bool {
 	return true // all is good, no state change needed
 }
 
-func (obj ServiceType) Apply() bool {
-	fmt.Printf("Apply->%v[%v]\n", obj.Type, obj.Name)
+func (obj *ServiceType) Apply() bool {
+	log.Printf("%v[%v]: Apply", obj.GetType(), obj.GetName())
 
 	if !util.IsRunningSystemd() {
 		log.Fatal("Systemd is not running.")
@@ -256,7 +282,7 @@ func (obj ServiceType) Apply() bool {
 		err = nil
 	}
 	if err != nil {
-		log.Printf("Unable to change startup status: %v\n", err)
+		log.Printf("Unable to change startup status: %v", err)
 		return false
 	}
 
@@ -292,3 +318,22 @@ func (obj ServiceType) Apply() bool {
 
 	return true
 }
+
+func (obj *ServiceType) Compare(typ Type) bool {
+	switch typ.(type) {
+	case *ServiceType:
+		typ := typ.(*ServiceType)
+		if obj.Name != typ.Name {
+			return false
+		}
+		if obj.State != typ.State {
+			return false
+		}
+		if obj.Startup != typ.Startup {
+			return false
+		}
+	default:
+		return false
+	}
+	return true
+}

tag.sh

@@ -0,0 +1,11 @@
+#!/bin/bash
+# TODO: don't run if current HEAD is already tagged (ensure this is idempotent)
+# take current HEAD with new version
+v=`git describe --match '[0-9]*\.[0-9]*\.[0-9]*' --tags --abbrev=0`
+t=`echo "${v%.*}.$((${v##*.}+1))"`	# increment version
+echo "Version $t is now tagged!"
+echo "Pushing $t to origin..."
+echo "Press ^C within 3s to abort."
+sleep 3s
+git tag $t
+git push origin $t

test.sh

@@ -2,6 +2,8 @@
 
 # test suite...
 echo running test.sh
+echo "ENV:"
+env
 
 # ensure there is no trailing whitespace or other whitespace errors
 git diff-tree --check $(git hash-object -t tree /dev/null) HEAD
@@ -9,3 +11,24 @@ git diff-tree --check $(git hash-object -t tree /dev/null) HEAD
 # ensure entries to authors file are sorted
 start=$(($(grep -n '^[[:space:]]*$' AUTHORS | awk -F ':' '{print $1}' | head -1) + 1))
 diff <(tail -n +$start AUTHORS | sort) <(tail -n +$start AUTHORS)
+
+./test/test-gofmt.sh
+./test/test-yamlfmt.sh
+./test/test-bashfmt.sh
+go test
+echo running go vet	# since it doesn't output an ok message on pass
+go vet && echo PASS
+
+# do these longer tests only when running on ci
+if env | grep -q -e '^TRAVIS=true$' -e '^JENKINS_URL=' -e '^BUILD_TAG=jenkins'; then
+	go test -race
+	./test/test-shell.sh
+else
+	# FIXME: this fails on travis for some reason
+	./test/test-reproducible.sh
+fi
+
+# run omv tests on jenkins physical hosts only
+if env | grep -q -e '^JENKINS_URL=' -e '^BUILD_TAG=jenkins'; then
+	./test/test-omv.sh
+fi

test/omv/helloworld.yaml

@@ -0,0 +1,52 @@
+---
+:domain: example.com
+:network: 192.168.123.0/24
+:image: fedora-23
+:cpus: ''
+:memory: ''
+:disks: 0
+:disksize: 40G
+:boxurlprefix: ''
+:sync: rsync
+:syncdir: ''
+:syncsrc: ''
+:folder: ".omv"
+:extern:
+- type: git
+  repository: https://github.com/purpleidea/mgmt
+  directory: mgmt
+:cd: ''
+:puppet: false
+:classes: []
+:shell:
+- mkdir /tmp/mgmt/
+:docker: false
+:kubernetes: false
+:ansible: []
+:playbook: []
+:ansible_extras: {}
+:cachier: false
+:vms:
+- :name: etcd
+  :shell:
+  - iptables -F
+  - cd /vagrant/mgmt/ && make path
+  - cd /vagrant/mgmt/ && make deps && make build && cp mgmt ~/bin/
+  - etcd -bind-addr "`hostname --ip-address`:2379" &
+  - cd && mgmt --help
+:namespace: omv
+:count: 0
+:username: ''
+:password: ''
+:poolid: true
+:repos: []
+:update: false
+:reboot: false
+:unsafe: false
+:nested: false
+:tests:
+- omv up etcd
+- vssh root@etcd -c pidof etcd
+- omv destroy
+:comment: simple hello world test case for mgmt
+:reallyrm: false

test/shell/.gitignore

@@ -0,0 +1 @@
+mgmt

test/shell/etcd.sh

@@ -0,0 +1,12 @@
+# NOTE: boiler plate to run etcd; source with: . etcd.sh; should NOT be +x
+cleanup ()
+{
+	killall etcd || killall -9 etcd || true	# kill etcd
+	rm -rf /tmp/etcd/
+}
+trap cleanup INT QUIT TERM EXIT ERR
+mkdir -p /tmp/etcd/
+cd /tmp/etcd/ >/dev/null	# shush the cd operation
+etcd &	# start etcd as job # 1
+sleep 1s	# let etcd startup
+cd - >/dev/null

test/shell/t1.sh

@@ -0,0 +1,14 @@
+#!/bin/bash
+# NOTES:
+#	* this is a simple shell based `mgmt` test case
+#	* it is recommended that you run mgmt wrapped in the timeout command
+#	* it is recommended that you run mgmt with --no-watch
+#	* it is recommended that you run mgmt --converged-timeout=<seconds>
+#	* you can run mgmt with --max-runtime=<seconds> in special scenarios
+#	* you can get a separate etcd going by sourcing etcd.sh: . etcd.sh
+
+set -o errexit
+set -o nounset
+set -o pipefail
+
+timeout --kill-after=3s 1s ./mgmt --help # hello world!

test/shell/t2.sh

@@ -0,0 +1,14 @@
+#!/bin/bash
+
+. etcd.sh	# start etcd as job # 1
+
+# run till completion
+timeout --kill-after=15s 10s ./mgmt run --file t2.yaml --converged-timeout=5 --no-watch &
+
+#jobs	# etcd is 1
+wait -n 2	# wait for mgmt to exit
+
+test -e /tmp/mgmt/f1
+test -e /tmp/mgmt/f2
+test -e /tmp/mgmt/f3
+test ! -e /tmp/mgmt/f4

test/shell/t2.yaml

@@ -0,0 +1,41 @@
+---
+graph: mygraph
+types:
+  noop:
+  - name: noop1
+  file:
+  - name: file1
+    path: "/tmp/mgmt/f1"
+    content: |
+      i am f1
+    state: exists
+  - name: file2
+    path: "/tmp/mgmt/f2"
+    content: |
+      i am f2
+    state: exists
+  - name: file3
+    path: "/tmp/mgmt/f3"
+    content: |
+      i am f3
+    state: exists
+  - name: file4
+    path: "/tmp/mgmt/f4"
+    content: |
+      i am f4 and i should not be here
+    state: absent
+edges:
+- name: e1
+  from:
+    type: file
+    name: file1
+  to:
+    type: file
+    name: file2
+- name: e2
+  from:
+    type: file
+    name: file2
+  to:
+    type: file
+    name: file3

test/shell/t3-a.yaml

@@ -0,0 +1,28 @@
+---
+graph: mygraph
+types:
+  file:
+  - name: file1a
+    path: "/tmp/mgmt/mgmtA/f1a"
+    content: |
+      i am f1
+    state: exists
+  - name: file2a
+    path: "/tmp/mgmt/mgmtA/f2a"
+    content: |
+      i am f2
+    state: exists
+  - name: "@@file3a"
+    path: "/tmp/mgmt/mgmtA/f3a"
+    content: |
+      i am f3, exported from host A
+    state: exists
+  - name: "@@file4a"
+    path: "/tmp/mgmt/mgmtA/f4a"
+    content: |
+      i am f4, exported from host A
+    state: exists
+collect:
+- type: file
+  pattern: "/tmp/mgmt/mgmtA/"
+edges: []

test/shell/t3-b.yaml

@@ -0,0 +1,28 @@
+---
+graph: mygraph
+types:
+  file:
+  - name: file1b
+    path: "/tmp/mgmt/mgmtB/f1b"
+    content: |
+      i am f1
+    state: exists
+  - name: file2b
+    path: "/tmp/mgmt/mgmtB/f2b"
+    content: |
+      i am f2
+    state: exists
+  - name: "@@file3b"
+    path: "/tmp/mgmt/mgmtB/f3b"
+    content: |
+      i am f3, exported from host B
+    state: exists
+  - name: "@@file4b"
+    path: "/tmp/mgmt/mgmtB/f4b"
+    content: |
+      i am f4, exported from host B
+    state: exists
+collect:
+- type: file
+  pattern: "/tmp/mgmt/mgmtB/"
+edges: []

test/shell/t3-c.yaml

@@ -0,0 +1,28 @@
+---
+graph: mygraph
+types:
+  file:
+  - name: file1c
+    path: "/tmp/mgmt/mgmtC/f1c"
+    content: |
+      i am f1
+    state: exists
+  - name: file2c
+    path: "/tmp/mgmt/mgmtC/f2c"
+    content: |
+      i am f2
+    state: exists
+  - name: "@@file3c"
+    path: "/tmp/mgmt/mgmtC/f3c"
+    content: |
+      i am f3, exported from host C
+    state: exists
+  - name: "@@file4c"
+    path: "/tmp/mgmt/mgmtC/f4c"
+    content: |
+      i am f4, exported from host C
+    state: exists
+collect:
+- type: file
+  pattern: "/tmp/mgmt/mgmtC/"
+edges: []

test/shell/t3.sh

@@ -0,0 +1,67 @@
+#!/bin/bash
+
+. etcd.sh	# start etcd as job # 1
+
+# setup
+mkdir -p "${MGMT_TMPDIR}"mgmt{A..C}
+
+# run till completion
+timeout --kill-after=15s 10s ./mgmt run --file t3-a.yaml --converged-timeout=5 --no-watch &
+timeout --kill-after=15s 10s ./mgmt run --file t3-b.yaml --converged-timeout=5 --no-watch &
+timeout --kill-after=15s 10s ./mgmt run --file t3-c.yaml --converged-timeout=5 --no-watch &
+
+. wait.sh	# wait for everything except etcd
+
+# A: collected
+test -e "${MGMT_TMPDIR}"mgmtA/f3b
+test -e "${MGMT_TMPDIR}"mgmtA/f3c
+test -e "${MGMT_TMPDIR}"mgmtA/f4b
+test -e "${MGMT_TMPDIR}"mgmtA/f4c
+
+# A: local
+test -e "${MGMT_TMPDIR}"mgmtA/f1a
+test -e "${MGMT_TMPDIR}"mgmtA/f2a
+test -e "${MGMT_TMPDIR}"mgmtA/f3a
+test -e "${MGMT_TMPDIR}"mgmtA/f4a
+
+# A: nope!
+test ! -e "${MGMT_TMPDIR}"mgmtA/f1b
+test ! -e "${MGMT_TMPDIR}"mgmtA/f2b
+test ! -e "${MGMT_TMPDIR}"mgmtA/f1c
+test ! -e "${MGMT_TMPDIR}"mgmtA/f2c
+
+# B: collected
+test -e "${MGMT_TMPDIR}"mgmtB/f3a
+test -e "${MGMT_TMPDIR}"mgmtB/f3c
+test -e "${MGMT_TMPDIR}"mgmtB/f4a
+test -e "${MGMT_TMPDIR}"mgmtB/f4c
+
+# B: local
+test -e "${MGMT_TMPDIR}"mgmtB/f1b
+test -e "${MGMT_TMPDIR}"mgmtB/f2b
+test -e "${MGMT_TMPDIR}"mgmtB/f3b
+test -e "${MGMT_TMPDIR}"mgmtB/f4b
+
+# B: nope!
+test ! -e "${MGMT_TMPDIR}"mgmtB/f1a
+test ! -e "${MGMT_TMPDIR}"mgmtB/f2a
+test ! -e "${MGMT_TMPDIR}"mgmtB/f1c
+test ! -e "${MGMT_TMPDIR}"mgmtB/f2c
+
+# C: collected
+test -e "${MGMT_TMPDIR}"mgmtC/f3a
+test -e "${MGMT_TMPDIR}"mgmtC/f3b
+test -e "${MGMT_TMPDIR}"mgmtC/f4a
+test -e "${MGMT_TMPDIR}"mgmtC/f4b
+
+# C: local
+test -e "${MGMT_TMPDIR}"mgmtC/f1c
+test -e "${MGMT_TMPDIR}"mgmtC/f2c
+test -e "${MGMT_TMPDIR}"mgmtC/f3c
+test -e "${MGMT_TMPDIR}"mgmtC/f4c
+
+# C: nope!
+test ! -e "${MGMT_TMPDIR}"mgmtC/f1a
+test ! -e "${MGMT_TMPDIR}"mgmtC/f2a
+test ! -e "${MGMT_TMPDIR}"mgmtC/f1b
+test ! -e "${MGMT_TMPDIR}"mgmtC/f2b

test/shell/t4.sh

@@ -0,0 +1,10 @@
+#!/bin/bash
+
+. etcd.sh	# start etcd as job # 1
+
+# should take slightly more than 25s, but fail if we take 35s)
+timeout --kill-after=35s 30s ./mgmt run --file t4.yaml --converged-timeout=5 --no-watch &
+
+#jobs	# etcd is 1
+#wait -n 2	# wait for mgmt to exit
+. wait.sh	# wait for everything except etcd

test/shell/t4.yaml

@@ -0,0 +1,77 @@
+---
+graph: mygraph
+comment: simple exec fan in example to demonstrate optimization)
+types:
+  exec:
+  - name: exec1
+    cmd: sleep 10s
+    shell: ''
+    timeout: 0
+    watchcmd: ''
+    watchshell: ''
+    ifcmd: ''
+    ifshell: ''
+    pollint: 0
+    state: present
+  - name: exec2
+    cmd: sleep 10s
+    shell: ''
+    timeout: 0
+    watchcmd: ''
+    watchshell: ''
+    ifcmd: ''
+    ifshell: ''
+    pollint: 0
+    state: present
+  - name: exec3
+    cmd: sleep 10s
+    shell: ''
+    timeout: 0
+    watchcmd: ''
+    watchshell: ''
+    ifcmd: ''
+    ifshell: ''
+    pollint: 0
+    state: present
+  - name: exec4
+    cmd: sleep 10s
+    shell: ''
+    timeout: 0
+    watchcmd: ''
+    watchshell: ''
+    ifcmd: ''
+    ifshell: ''
+    pollint: 0
+    state: present
+  - name: exec5
+    cmd: sleep 10s
+    shell: ''
+    timeout: 0
+    watchcmd: ''
+    watchshell: ''
+    ifcmd: ''
+    ifshell: ''
+    pollint: 0
+    state: present
+edges:
+- name: e1
+  from:
+    type: exec
+    name: exec1
+  to:
+    type: exec
+    name: exec5
+- name: e2
+  from:
+    type: exec
+    name: exec2
+  to:
+    type: exec
+    name: exec5
+- name: e3
+  from:
+    type: exec
+    name: exec3
+  to:
+    type: exec
+    name: exec5

test/shell/t5.sh

@@ -0,0 +1,10 @@
+#!/bin/bash
+
+. etcd.sh	# start etcd as job # 1
+
+# should take slightly more than 35s, but fail if we take 45s)
+timeout --kill-after=45s 40s ./mgmt run --file t5.yaml --converged-timeout=5 --no-watch &
+
+#jobs	# etcd is 1
+#wait -n 2	# wait for mgmt to exit
+. wait.sh	# wait for everything except etcd

test/shell/t5.yaml

@@ -0,0 +1,128 @@
+---
+graph: mygraph
+comment: simple exec fan in to fan out example to demonstrate optimization
+types:
+  exec:
+  - name: exec1
+    cmd: sleep 10s
+    shell: ''
+    timeout: 0
+    watchcmd: ''
+    watchshell: ''
+    ifcmd: ''
+    ifshell: ''
+    pollint: 0
+    state: present
+  - name: exec2
+    cmd: sleep 10s
+    shell: ''
+    timeout: 0
+    watchcmd: ''
+    watchshell: ''
+    ifcmd: ''
+    ifshell: ''
+    pollint: 0
+    state: present
+  - name: exec3
+    cmd: sleep 10s
+    shell: ''
+    timeout: 0
+    watchcmd: ''
+    watchshell: ''
+    ifcmd: ''
+    ifshell: ''
+    pollint: 0
+    state: present
+  - name: exec4
+    cmd: sleep 10s
+    shell: ''
+    timeout: 0
+    watchcmd: ''
+    watchshell: ''
+    ifcmd: ''
+    ifshell: ''
+    pollint: 0
+    state: present
+  - name: exec5
+    cmd: sleep 10s
+    shell: ''
+    timeout: 0
+    watchcmd: ''
+    watchshell: ''
+    ifcmd: ''
+    ifshell: ''
+    pollint: 0
+    state: present
+  - name: exec6
+    cmd: sleep 10s
+    shell: ''
+    timeout: 0
+    watchcmd: ''
+    watchshell: ''
+    ifcmd: ''
+    ifshell: ''
+    pollint: 0
+    state: present
+  - name: exec7
+    cmd: sleep 10s
+    shell: ''
+    timeout: 0
+    watchcmd: ''
+    watchshell: ''
+    ifcmd: ''
+    ifshell: ''
+    pollint: 0
+    state: present
+  - name: exec8
+    cmd: sleep 15s
+    shell: ''
+    timeout: 0
+    watchcmd: ''
+    watchshell: ''
+    ifcmd: ''
+    ifshell: ''
+    pollint: 0
+    state: present
+edges:
+- name: e1
+  from:
+    type: exec
+    name: exec1
+  to:
+    type: exec
+    name: exec4
+- name: e2
+  from:
+    type: exec
+    name: exec2
+  to:
+    type: exec
+    name: exec4
+- name: e3
+  from:
+    type: exec
+    name: exec3
+  to:
+    type: exec
+    name: exec4
+- name: e4
+  from:
+    type: exec
+    name: exec4
+  to:
+    type: exec
+    name: exec5
+- name: e5
+  from:
+    type: exec
+    name: exec4
+  to:
+    type: exec
+    name: exec6
+- name: e6
+  from:
+    type: exec
+    name: exec4
+  to:
+    type: exec
+    name: exec7

test/shell/wait.sh

@@ -0,0 +1,6 @@
+# NOTE: boiler plate to wait on mgmt; source with: . wait.sh; should NOT be +x
+for j in `jobs -p`
+do
+	[ $j -eq `pidof etcd` ] && continue	# don't wait for etcd
+	wait $j	# wait for mgmt job $j
+done

test/test-bashfmt.sh

@@ -0,0 +1,31 @@
+#!/bin/bash
+# check for any bash files that aren't properly formatted
+# TODO: this is hardly exhaustive
+
+set -o errexit
+set -o nounset
+set -o pipefail
+
+ROOT=$(dirname "${BASH_SOURCE}")/..
+
+cd "${ROOT}"
+
+find_files() {
+	git ls-files | grep -e '\.sh$' -e '\.bash$'
+}
+
+bad_files=$(
+	for i in $(find_files); do
+		# search for more than one leading space, to ensure we use tabs
+		if grep -q '^  ' "$i"; then
+			echo "$i"
+		fi
+	done
+)
+
+if [[ -n "${bad_files}" ]]; then
+	echo 'FAIL'
+	echo 'The following bash files are not properly formatted:'
+	echo "${bad_files}"
+	exit 1
+fi

test/test-gofmt.sh

@@ -9,7 +9,7 @@ ROOT=$(dirname "${BASH_SOURCE}")/..
 
 GO_VERSION=($(go version))
 
-if [[ -z $(echo "${GO_VERSION[2]}" | grep -E 'go1.2|go1.3|go1.4') ]]; then
+if [[ -z $(echo "${GO_VERSION[2]}" | grep -E 'go1.2|go1.3|go1.4|go1.5') ]]; then
 	echo "Unknown go version '${GO_VERSION}', skipping gofmt."
 	exit 0
 fi
@@ -17,19 +17,14 @@ fi
 cd "${ROOT}"
 
 find_files() {
-	find . -not \( \
-		\( \
-			-wholename './old' \
-			-o -wholename './tmp' \
-		\) -prune \
-	\) -name '*.go'
+	git ls-files | grep '\.go$'
 }
 
 GOFMT="gofmt"	# we prefer to not use the -s flag, which is pretty annoying...
 bad_files=$(find_files | xargs $GOFMT -l)
 if [[ -n "${bad_files}" ]]; then
 	echo 'FAIL'
-	echo 'The following files are not properly formatted:'
+	echo 'The following golang files are not properly formatted:'
 	echo "${bad_files}"
 	exit 1
 fi

test/test-omv.sh

@@ -0,0 +1,11 @@
+#!/bin/bash -ie
+# simple test harness for testing mgmt via omv
+CWD=`pwd`
+DIR="$( cd "$( dirname "${BASH_SOURCE[0]}" )" && pwd )"	# dir!
+cd "$DIR" >/dev/null	# work from test directory
+
+# vtest+ tests
+vtest+ omv/helloworld.yaml
+
+# return to original dir
+cd "$CWD" >/dev/null

test/test-reproducible.sh

@@ -0,0 +1,39 @@
+#!/bin/bash
+# simple test for reproducibility, probably needs major improvements
+
+set -o errexit
+set -o pipefail
+
+DIR="$( cd "$( dirname "${BASH_SOURCE[0]}" )" && cd .. && pwd )"	# dir!
+cd "$DIR" >/dev/null	# work from main mgmt directory
+make build
+T=`mktemp --tmpdir -d tmp.XXX`
+cp -a ./mgmt "$T"/mgmt.1
+make clean
+make build
+cp -a ./mgmt "$T"/mgmt.2
+
+# size comparison test
+[ `stat -c '%s' "$T"/mgmt.1` -eq `stat -c '%s' "$T"/mgmt.2` ] || failures="Size of binary was not reproducible"
+
+# sha1sum test
+sha1sum "$T"/mgmt.1 > "$T"/mgmt.SHA1SUMS.1
+sha1sum "$T"/mgmt.2 > "$T"/mgmt.SHA1SUMS.2
+cat "$T"/mgmt.SHA1SUMS.1 | sed 's/mgmt\.1/mgmt\.X/' > "$T"/mgmt.SHA1SUMS.1X
+cat "$T"/mgmt.SHA1SUMS.2 | sed 's/mgmt\.2/mgmt\.X/' > "$T"/mgmt.SHA1SUMS.2X
+diff -q "$T"/mgmt.SHA1SUMS.1X "$T"/mgmt.SHA1SUMS.2X || failures=$( [ -n "${failures}" ] && echo "$failures" ; echo "SHA1SUM of binary was not reproducible" )
+
+# clean up
+if [ "$T" != '' ]; then
+	rm -rf "$T"
+fi
+make clean
+
+# display errors
+if [[ -n "${failures}" ]]; then
+	echo 'FAIL'
+	echo 'The following tests failed:'
+	echo "${failures}"
+	exit 1
+fi
+echo PASS

test/test-shell.sh

@@ -0,0 +1,73 @@
+#!/bin/bash
+# simple test harness for testing mgmt
+# NOTE: this will rm -rf /tmp/mgmt/
+
+set -o errexit
+set -o pipefail
+
+LINE=$(printf '=%.0s' `seq -s ' ' $(tput cols)`)	# a terminal width string
+DIR="$( cd "$( dirname "${BASH_SOURCE[0]}" )" && cd .. && pwd )"	# dir!
+cd "$DIR" >/dev/null	# work from main mgmt directory
+make build
+MGMT="$DIR/test/shell/mgmt"
+cp -a "$DIR/mgmt" "$MGMT"	# put a copy there
+failures=""
+count=0
+
+# loop through tests
+for i in $DIR/test/shell/*.sh; do
+	[ -x "$i" ] || continue	# file must be executable
+	ii=`basename "$i"`	# short name
+	# if ARGV has test names, only execute those!
+	if [ "$1" != '' ]; then
+		[ "$ii" != "$1" ] && continue
+	fi
+	cd $DIR/test/shell/ >/dev/null	# shush the cd operation
+	mkdir -p '/tmp/mgmt/'	# directory for mgmt to put files in
+	#echo "Running: $ii"
+	export MGMT_TMPDIR='/tmp/mgmt/'	# we can add to env like this
+	count=`expr $count + 1`
+	set +o errexit	# don't kill script on test failure
+	out=$($i 2>&1)	# run and capture stdout & stderr
+	e=$?	# save exit code
+	set -o errexit	# re-enable killing on script failure
+	cd - >/dev/null
+	rm -rf '/tmp/mgmt/'	# clean up after test
+	if [ $e -ne 0 ]; then
+		# store failures...
+		failures=$(
+			# prepend previous failures if any
+			[ -n "${failures}" ] && echo "$failures" && echo "$LINE"
+			echo "Script: $ii"
+			# if we see 124, it might be the exit value of timeout!
+			[ $e -eq 124 ] && echo "Exited: $e (timeout?)" || echo "Exited: $e"
+			if [ "$out" = "" ]; then
+				echo "Output: (empty!)"
+			else
+				echo "Output:"
+				echo "$out"
+			fi
+		)
+	else
+		echo -e "ok\t$ii"	# pass
+	fi
+done
+
+# clean up
+rm -f "$MGMT"
+make clean
+
+if [ "$count" = '0' ]; then
+	echo 'FAIL'
+	echo 'No tests were run!'
+	exit 1
+fi
+
+# display errors
+if [[ -n "${failures}" ]]; then
+	echo 'FAIL'
+	echo 'The following tests failed:'
+	echo "${failures}"
+	exit 1
+fi
+echo PASS

test/test-yamlfmt.sh

@@ -5,21 +5,33 @@ set -o errexit
 set -o nounset
 set -o pipefail
 
+if env | grep -q -e '^TRAVIS=true$' -e '^JENKINS_URL=' -e '^BUILD_TAG=jenkins'; then
+	echo "Travis and Jenkins give wonky results here, skipping test!"
+	exit 0
+fi
+
 ROOT=$(dirname "${BASH_SOURCE}")/..
+
+RUBY=`which ruby 2>/dev/null`
+if [ -z $RUBY ]; then
+	echo "The 'ruby' utility can't be found."
+	exit 1
+fi
+
+$RUBY -e "require 'yaml'" 2>/dev/null || (
+	echo "The ruby 'yaml' library can't be found."
+	exit 1
+)
+
 cd "${ROOT}"
 
 find_files() {
-	find . -not \( \
-		\( \
-			-wholename './old' \
-			-o -wholename './tmp' \
-		\) -prune \
-	\) -name '*.yaml'
+	git ls-files | grep '\.yaml$'
 }
 
 bad_files=$(
 	for i in $(find_files); do
-		if ! diff -q <( ruby -e "require 'yaml'; puts YAML.load_file('$i').to_yaml" 2>/dev/null ) <( cat "$i" ) &>/dev/null; then
+		if ! diff -q <( ruby -e "require 'yaml'; puts YAML.load_file('$i').to_yaml.each_line.map(&:rstrip).join(10.chr)+10.chr" 2>/dev/null ) <( cat "$i" ) &>/dev/null; then
 			echo "$i"
 		fi
 	done
@@ -27,7 +39,7 @@ bad_files=$(
 
 if [[ -n "${bad_files}" ]]; then
 	echo 'FAIL'
-	echo 'The following files are not properly formatted:'
+	echo 'The following yaml files are not properly formatted:'
 	echo "${bad_files}"
 	exit 1
 fi

types.go

@@ -1,5 +1,5 @@
 // Mgmt
-// Copyright (C) 2013-2015+ James Shubin and the project contributors
+// Copyright (C) 2013-2016+ James Shubin and the project contributors
 // Written by James Shubin <james@shubin.ca> and the project contributors
 //
 // This program is free software: you can redistribute it and/or modify
@@ -18,56 +18,390 @@
 package main
 
 import (
-	"code.google.com/p/go-uuid/uuid"
-	"fmt"
 	"log"
+	"time"
+)
+
+//go:generate stringer -type=typeState -output=typestate_stringer.go
+type typeState int
+
+const (
+	typeNil typeState = iota
+	typeWatching
+	typeEvent // an event has happened, but we haven't poked yet
+	typeApplying
+	typePoking
+)
+
+//go:generate stringer -type=typeConvergedState -output=typeconvergedstate_stringer.go
+type typeConvergedState int
+
+const (
+	typeConvergedNil typeConvergedState = iota
+	//typeConverged
+	typeConvergedTimeout
 )
 
 type Type interface {
-	//Name()  string
-	Watch(*Vertex)
+	Init()
+	GetName() string // can't be named "Name()" because of struct field
+	GetType() string
+	Watch()
 	StateOK() bool // TODO: can we rename this to something better?
 	Apply() bool
-	Exit() bool
+	SetVertex(*Vertex)
+	SetConvegedCallback(ctimeout int, converged chan bool)
+	Compare(Type) bool
+	SendEvent(eventName, bool, bool) bool
+	IsWatching() bool
+	SetWatching(bool)
+	GetConvergedState() typeConvergedState
+	SetConvergedState(typeConvergedState)
+	GetState() typeState
+	SetState(typeState)
+	GetTimestamp() int64
+	UpdateTimestamp() int64
+	OKTimestamp() bool
+	Poke(bool)
+	BackPoke()
+}
+
+type BaseType struct {
+	Name           string `yaml:"name"`
+	timestamp      int64  // last updated timestamp ?
+	events         chan Event
+	vertex         *Vertex
+	state          typeState
+	convergedState typeConvergedState
+	watching       bool // is Watch() loop running ?
+	ctimeout       int  // converged timeout
+	converged      chan bool
+	isStateOK      bool // whether the state is okay based on events or not
 }
 
 type NoopType struct {
-	uuid   string
-	Type   string      // always "noop"
-	Name   string      // name variable
-	Events chan string // FIXME: eventually a struct for the event?
+	BaseType `yaml:",inline"`
+	Comment  string `yaml:"comment"` // extra field for example purposes
 }
 
 func NewNoopType(name string) *NoopType {
+	// FIXME: we could get rid of this New constructor and use raw object creation with a required Init()
 	return &NoopType{
-		uuid:   uuid.New(),
-		Type:   "noop",
-		Name:   name,
-		Events: make(chan string, 1), // XXX: chan size?
+		BaseType: BaseType{
+			Name:   name,
+			events: make(chan Event), // unbuffered chan size to avoid stale events
+			vertex: nil,
+		},
+		Comment: "",
 	}
 }
 
-func (obj NoopType) Watch(v *Vertex) {
-	select {
-	case exit := <-obj.Events:
-		if exit == "exit" {
-			return
+// initialize structures like channels if created without New constructor
+func (obj *BaseType) Init() {
+	obj.events = make(chan Event)
+}
+
+// this method gets used by all the types, if we have one of (obj NoopType) it would get overridden in that case!
+func (obj *BaseType) GetName() string {
+	return obj.Name
+}
+
+func (obj *BaseType) GetType() string {
+	return "Base"
+}
+
+func (obj *BaseType) GetVertex() *Vertex {
+	return obj.vertex
+}
+
+func (obj *BaseType) SetVertex(v *Vertex) {
+	obj.vertex = v
+}
+
+func (obj *BaseType) SetConvegedCallback(ctimeout int, converged chan bool) {
+	obj.ctimeout = ctimeout
+	obj.converged = converged
+}
+
+// is the Watch() function running?
+func (obj *BaseType) IsWatching() bool {
+	return obj.watching
+}
+
+// store status of if the Watch() function is running
+func (obj *BaseType) SetWatching(b bool) {
+	obj.watching = b
+}
+
+func (obj *BaseType) GetConvergedState() typeConvergedState {
+	return obj.convergedState
+}
+
+func (obj *BaseType) SetConvergedState(state typeConvergedState) {
+	obj.convergedState = state
+}
+
+func (obj *BaseType) GetState() typeState {
+	return obj.state
+}
+
+func (obj *BaseType) SetState(state typeState) {
+	if DEBUG {
+		log.Printf("%v[%v]: State: %v -> %v", obj.GetType(), obj.GetName(), obj.GetState(), state)
+	}
+	obj.state = state
+}
+
+// GetTimestamp returns the timestamp of a vertex
+func (obj *BaseType) GetTimestamp() int64 {
+	return obj.timestamp
+}
+
+// UpdateTimestamp updates the timestamp on a vertex and returns the new value
+func (obj *BaseType) UpdateTimestamp() int64 {
+	obj.timestamp = time.Now().UnixNano() // update
+	return obj.timestamp
+}
+
+// can this element run right now?
+func (obj *BaseType) OKTimestamp() bool {
+	v := obj.GetVertex()
+	g := v.GetGraph()
+	// these are all the vertices pointing TO v, eg: ??? -> v
+	for _, n := range g.IncomingGraphEdges(v) {
+		// if the vertex has a greater timestamp than any pre-req (n)
+		// then we can't run right now...
+		// if they're equal (eg: on init of 0) then we also can't run
+		// b/c we should let our pre-req's go first...
+		x, y := obj.GetTimestamp(), n.Type.GetTimestamp()
+		if DEBUG {
+			log.Printf("%v[%v]: OKTimestamp: (%v) >= %v[%v](%v): !%v", obj.GetType(), obj.GetName(), x, n.GetType(), n.GetName(), y, x >= y)
+		}
+		if x >= y {
+			return false
+		}
+	}
+	return true
+}
+
+// notify nodes after me in the dependency graph that they need refreshing...
+// NOTE: this assumes that this can never fail or need to be rescheduled
+func (obj *BaseType) Poke(activity bool) {
+	v := obj.GetVertex()
+	g := v.GetGraph()
+	// these are all the vertices pointing AWAY FROM v, eg: v -> ???
+	for _, n := range g.OutgoingGraphEdges(v) {
+		// XXX: if we're in state event and haven't been cancelled by
+		// apply, then we can cancel a poke to a child, right? XXX
+		// XXX: if n.Type.GetState() != typeEvent { // is this correct?
+		if true { // XXX
+			if DEBUG {
+				log.Printf("%v[%v]: Poke: %v[%v]", v.GetType(), v.GetName(), n.GetType(), n.GetName())
+			}
+			n.SendEvent(eventPoke, false, activity) // XXX: can this be switched to sync?
 		} else {
-			log.Fatal("Unknown event: %v\n", exit)
+			if DEBUG {
+				log.Printf("%v[%v]: Poke: %v[%v]: Skipped!", v.GetType(), v.GetName(), n.GetType(), n.GetName())
+			}
 		}
 	}
 }
 
-func (obj NoopType) Exit() bool {
-	obj.Events <- "exit"
+// poke the pre-requisites that are stale and need to run before I can run...
+func (obj *BaseType) BackPoke() {
+	v := obj.GetVertex()
+	g := v.GetGraph()
+	// these are all the vertices pointing TO v, eg: ??? -> v
+	for _, n := range g.IncomingGraphEdges(v) {
+		x, y, s := obj.GetTimestamp(), n.Type.GetTimestamp(), n.Type.GetState()
+		// if the parent timestamp needs poking AND it's not in state
+		// typeEvent, then poke it. If the parent is in typeEvent it
+		// means that an event is pending, so we'll be expecting a poke
+		// back soon, so we can safely discard the extra parent poke...
+		// TODO: implement a stateLT (less than) to tell if something
+		// happens earlier in the state cycle and that doesn't wrap nil
+		if x >= y && (s != typeEvent && s != typeApplying) {
+			if DEBUG {
+				log.Printf("%v[%v]: BackPoke: %v[%v]", v.GetType(), v.GetName(), n.GetType(), n.GetName())
+			}
+			n.SendEvent(eventBackPoke, false, false) // XXX: can this be switched to sync?
+		} else {
+			if DEBUG {
+				log.Printf("%v[%v]: BackPoke: %v[%v]: Skipped!", v.GetType(), v.GetName(), n.GetType(), n.GetName())
+			}
+		}
+	}
+}
+
+// push an event into the message queue for a particular type vertex
+func (obj *BaseType) SendEvent(event eventName, sync bool, activity bool) bool {
+	// TODO: isn't this race-y ?
+	if !obj.IsWatching() { // element has already exited
+		return false // if we don't return, we'll block on the send
+	}
+	if !sync {
+		obj.events <- Event{event, nil, "", activity}
+		return true
+	}
+
+	resp := make(chan bool)
+	obj.events <- Event{event, resp, "", activity}
+	for {
+		value := <-resp
+		// wait until true value
+		if value {
+			return true
+		}
+	}
+}
+
+// process events when a select gets one, this handles the pause code too!
+// the return values specify if we should exit and poke respectively
+func (obj *BaseType) ReadEvent(event *Event) (exit, poke bool) {
+	event.ACK()
+	switch event.Name {
+	case eventStart:
+		return false, true
+
+	case eventPoke:
+		return false, true
+
+	case eventBackPoke:
+		return false, true // forward poking in response to a back poke!
+
+	case eventExit:
+		return true, false
+
+	case eventPause:
+		// wait for next event to continue
+		select {
+		case e := <-obj.events:
+			e.ACK()
+			if e.Name == eventExit {
+				return true, false
+			} else if e.Name == eventStart { // eventContinue
+				return false, false // don't poke on unpause!
+			} else {
+				// if we get a poke event here, it's a bug!
+				log.Fatalf("%v[%v]: Unknown event: %v, while paused!", obj.GetType(), obj.GetName(), e)
+			}
+		}
+
+	default:
+		log.Fatal("Unknown event: ", event)
+	}
+	return true, false // required to keep the stupid go compiler happy
+}
+
+// useful for using as: return CleanState() in the StateOK functions when there
+// are multiple `true` return exits
+func (obj *BaseType) CleanState() bool {
+	obj.isStateOK = true
 	return true
 }
 
-func (obj NoopType) StateOK() bool {
+// XXX: rename this function
+func Process(obj Type) {
+	if DEBUG {
+		log.Printf("%v[%v]: Process()", obj.GetType(), obj.GetName())
+	}
+	obj.SetState(typeEvent)
+	var ok = true
+	var apply = false // did we run an apply?
+	// is it okay to run dependency wise right now?
+	// if not, that's okay because when the dependency runs, it will poke
+	// us back and we will run if needed then!
+	if obj.OKTimestamp() {
+		if DEBUG {
+			log.Printf("%v[%v]: OKTimestamp(%v)", obj.GetType(), obj.GetName(), obj.GetTimestamp())
+		}
+		if !obj.StateOK() { // TODO: can we rename this to something better?
+			if DEBUG {
+				log.Printf("%v[%v]: !StateOK()", obj.GetType(), obj.GetName())
+			}
+			// throw an error if apply fails...
+			// if this fails, don't UpdateTimestamp()
+			obj.SetState(typeApplying)
+			if !obj.Apply() { // check for error
+				ok = false
+			} else {
+				apply = true
+			}
+		}
+
+		if ok {
+			// update this timestamp *before* we poke or the poked
+			// nodes might fail due to having a too old timestamp!
+			obj.UpdateTimestamp()    // this was touched...
+			obj.SetState(typePoking) // can't cancel parent poke
+			obj.Poke(apply)
+		}
+		// poke at our pre-req's instead since they need to refresh/run...
+	} else {
+		// only poke at the pre-req's that need to run
+		go obj.BackPoke()
+	}
+}
+
+func (obj *NoopType) GetType() string {
+	return "Noop"
+}
+
+func (obj *NoopType) Watch() {
+	if obj.IsWatching() {
+		return
+	}
+	obj.SetWatching(true)
+	defer obj.SetWatching(false)
+
+	//vertex := obj.vertex // stored with SetVertex
+	var send = false // send event?
+	var exit = false
+	for {
+		obj.SetState(typeWatching) // reset
+		select {
+		case event := <-obj.events:
+			obj.SetConvergedState(typeConvergedNil)
+			// we avoid sending events on unpause
+			if exit, send = obj.ReadEvent(&event); exit {
+				return // exit
+			}
+
+		case _ = <-TimeAfterOrBlock(obj.ctimeout):
+			obj.SetConvergedState(typeConvergedTimeout)
+			obj.converged <- true
+			continue
+		}
+
+		// do all our event sending all together to avoid duplicate msgs
+		if send {
+			send = false
+			// only do this on certain types of events
+			//obj.isStateOK = false // something made state dirty
+			Process(obj) // XXX: rename this function
+		}
+	}
+}
+
+func (obj *NoopType) StateOK() bool {
 	return true // never needs updating
 }
 
-func (obj NoopType) Apply() bool {
-	fmt.Printf("Apply->%v[%v]\n", obj.Type, obj.Name)
+func (obj *NoopType) Apply() bool {
+	log.Printf("%v[%v]: Apply", obj.GetType(), obj.GetName())
+	return true
+}
+
+func (obj *NoopType) Compare(typ Type) bool {
+	switch typ.(type) {
+	// we can only compare NoopType to others of the same type
+	case *NoopType:
+		typ := typ.(*NoopType)
+		if obj.Name != typ.Name {
+			return false
+		}
+	default:
+		return false
+	}
 	return true
 }