2c40b8734e
Add a utility that folds an iterator into a result and uses it to reduce boilerplate in the codebase. |
||
---|---|---|
bin | ||
doc | ||
imag-counter | ||
imag-diary | ||
imag-link | ||
imag-notes | ||
imag-ref | ||
imag-store | ||
imag-tag | ||
imag-view | ||
libimagcounter | ||
libimagdiary | ||
libimagentryfilter | ||
libimagentrylink | ||
libimagentrylist | ||
libimagentrymarkdown | ||
libimagentryselect | ||
libimagentrytag | ||
libimagentryview | ||
libimagerror | ||
libimaginteraction | ||
libimagnotes | ||
libimagref | ||
libimagrt | ||
libimagstore | ||
libimagstorestdhook | ||
libimagtimeui | ||
libimagutil | ||
tests | ||
.editorconfig | ||
.gitignore | ||
.travis.yml | ||
CONTRIBUTING.md | ||
default.nix | ||
imagrc.toml | ||
LICENSE | ||
Makefile | ||
README.md |
imag
Imag is a CLI PIM suite with a nice API-ish commandline interface, so you can integrate it in your tools of choice (Editor, MUA, RSS reader, etc etc).
Goal
Our goal is to
Create a fast, reliable, forwards/backwards compatible commandline personal information management suite which covers all aspects of personal information management, consists of reusable parts and integrates well with known commandline tools.
We try to accomplish these requirements:
- "fast": We use the awesome, fast and safe programming language "Rust"
- "reliable": We try to test every aspect of our software. Our build process ensures that the build breaks whenever a library interface changes and the modules which use the library are not updated.
- "forwards/backwards compatible:" Our (plain text) on-disk data format and storage library both ensure that incompatibilities are captured and resolved (using semver)
- "commandline": We ensure that everything can be done by commandline calls, for some modules there might be a curses-like UI, but there are no graphical clients and there never will be any within this codebase. We use clap for commandline-interface building and we try to keep the interface easy and consistent between modules.
- "personal": We store everything as plain text in a store inside the users
$HOME
directory. There will be a version-control (most surelygit
) hook integrated to sync between several machines. There are no multi-user features included or planned at the time of writing. - "information management": We want to give the user the possibility to put every single information about their personal lives into the store and we try hard to provide a sane interface to query and retrieve data from this database.
- "covers all the aspects of personal information management": We want to
provide modules for:
- contact management
- calendar
- diary
- notes
- personal wiki
- news (rss)
- passwords
- images
- music
- movies
- personal project management
- podcast management
- ledger
- bibliography management
- ... and many, many more.
- "constists of reusable parts": Every functionality is implemented as library. The binaries we ship are just commandline-interace-to-library-interface translators
- "integrates well with known commandline tools": We do not re-invent the wheel.
We do not implement "yet another password manager", but use
the standard unix password manager, do not
implement a news reader, but use newsbeuter,
do not reimplement a mail reader, etc etc.
We do not copy images, movies or other data to the store but "link" them into
the store, so you can use imag tools to query and access this data, but still
live with your beloved commandline apps. We do not want to duplicate work but
reuse as much as possible.
You don't like one of the applications we use (for example
pass
as password manager)? Sure, feel free to submit patches so the user is able to switch the used tool, as long as it doesn't break the workflow. We will happily merge them!
Current state of development
This application is in really early development.
We have implemented the very core of the system: the store library. There's also some progress on utility libraries for linking entries, tagging and the hook system of the store is in progress as well. There is also one commandline application: "imag-store" (the "store" subcommand) available by now, but this is meant for developers and debugging purposes as it provides direct core-level store access.
Though, the very core of the system is stable and nothing prevents you from contributing and implementing a module, though some convenience is not yet provided (as the libraries are work-in-progress).
Building/Running
Building
One can build all the modules simply by running make
which defaults to building all the modules
and placing them in the out/
directory of the project root.
$> make
...
$> ls out/
imag-counter imag-link imag-notes imag-store imag-tag imag-view
Building all the modules may take some time, so alternatively one can build only a specific module
by runing $> make $module
where $module
is one of the imag-*
names, such as imag-counter
,
imag-link
, etc.
Running
To run imag, simply call ./bin/imag
. This script has a function to search for
modules, which utilizes an environment variable called IMAG_IS_THE_SHIT
.
To run imag with all components:
$> IMAG_IS_THE_SHIT=$(pwd) ./bin/imag
assuming you are currently in the imag source directory. Otherwise it is
$>IMAG_IS_THE_SHIT=$IMAG_SRC $IMAG_SRC/bin/imag
if $IMAG_SOURCE points to the imag source directory.
To test, simply add --help
to one of the above commands. Imag should now be
able to find the available commands.
Documentation
For detailed information, please read the documentation (You can either read the Markdown files or compile it to HTML/PDF using pandoc).
License
We chose to distribute this software under terms of GNU LGPLv2.1.
This decision was made to ensure everyone can write applications which use the imag core functionality which is distributed with the imag source distribution.