readme

• Preliminaries
• Prerequisites
  • bfg
  • Dependencies:
• How to build
• How to use
• License
• TODO

Preliminaries

• Code is written in scala, C++ and perl.
• For scala code, use sbt to build and run programs. It is the easiest
• This code has only being tested under Linux, but it should run under MacOS

Prerequisites

srcml	https://www.srcml.org/	Make sure srcml is in path
ctags	https://github.com/universal-ctags	Make sure ctags is in path
bfg	https://github.com/dmgerman/bfg-repo-cleaner/tree/blobexec	Download this branch

bfg

• obtain the above bfg branch

git clone https://github.com/dmgerman/bfg-repo-cleaner.git --branch blobexec

• build in the bfg root directory

sbt
bfg/assembly

Dependencies:

For each module, its dependencies are documented in their corresponding readme.org file.

As an example, on Debian 9 the following packages must be installed: cmake libarchive-dev libxml++2.6-dev libxml2-dev libcurl4-openssl-dev libxslt1-dev libboost-all-dev libantlr-dev libssl-dev libxerces-c-dev exuberant-ctags libdbi-perl libjgit-java libhtml-fromtext-perl libset-scalar-perl libdbd-sqlite3-perl

How to build

• compile slickGitLog, persons, remapCommits with sbt in their directories (sbt can be found at

https://www.scala-sbt.org/download.html repo)

sbt one-jar

• use make to compile srcMLtoken
• perl scripts can be run without compilation

How to use

This is the workflow to process a git repository with cregit, and to generate the HTML views of its contributions.

• It assumes that the cregit-repository will be created in /tmp/xournal
• The original repository is located at /home/dmg/git.hacking.mine/xournal

1. Create the view repository.

To tokenize the files we require to create some environment variables to communicate with the tokenizing script

BFG_MEMO_DIR	directory to use for memoization of tokenized files
BFG_TOKENIZE_CMD	command to use to tokenize, might include parameters

Example:

• use the tokenizeByBlobId/tokenBySha.pl to do the tokenization
• tokenBySha.pl will execute

./tokenizeSrcMl.pl --srcml2token=<path to srcml2token> --srcml=<path to srcml> --ctags=<path to ctags>

Run it as:

export BFG_MEMO_DIR=/tmp/memo
export BFG_TOKENIZE_CMD=/home/dmg/git.dmg/cregit-scala/tokenize/tokenizeSrcMl.pl --srcml2token=/home/dmg/git.dmg/cregit-scala/tokenize/srcMLtoken/srcml2token --srcml=srcml --ctags=/usr/local/bin/ctags
java -jar bfg-cregit.jar '--blob-exec:/home/dmg/git.dmg/cregit-scala/tokenizeByBlobId/tokenBySha.pl=.[ch]$' --no-blob-protection /path/repo

1. Create the history database for the original repo

java -jar slickGitLog.jar /tmp/xournal-original.db /home/dmg/git.hacking.mine/xournal

1. Create the history database for the cregit repo

java -jar slickGitLog.jar /tmp/xournal-cregit.db /tmp/xournal

1. Create the persons database

java -jar persons.jar /home/dmg/git.hacking.mine/xournal /tmp/xournal.xls /tmp/xournal-persons.db

1. Create blame of cregit files

perl blameRepoFiles.pl --verbose --formatBlame=./formatBlame.pl /tmp/xournal /tmp/blame '\.[ch]$'

1. create the table with the map from newcommits to commits

java -jar remapCommits.jar /tmp/xournal-cregit.db /tmp/xournal

1. Create html version of the files

Example:

perl ./prettyPrintFiles.pl --verbose /tmp/xournal-cregit.db /tmp/xournal-persons.db ~/git.hacking.mine/xournal /tmp/blame /tmp/html https://github.com/dmgerman/xournal-next/commit/ '\.[ch]$'

License

The license of Cregit is GPL-3.0+

TODO

• create temporary files in TMP proper location
• use preferred name in html files
• create a driver program for processing an entire repository
• customize programs to read a JSON file with configuration?