maindocoll - Summary

 
 
Membership Info
Group Admin:
1 active member

Group identification
Id: 10893
System Name: docoll
Name: docoll
Group Type: non-GNU software and documentation

Search in this Group


 

This group is not part of the GNU Project.

The problem

An organisation had produced many documents as computer files but had no document repository; the documents existed on the hard drives of  Windows and GNU-Linux personal computers.  There were many duplicate files, sometimes with different names.  Documents had been copied without preserving the last modification times.

Sometimes a document was known to exist and was required urgently; significant staff time was used looking for it.  Sometimes documents could not be found and were re-typed from hard copy.

There was no text searchable collation so people had no mechanism to search for prior relevant works. If long-serving people did not recall prior relevant documents, similar work could be repeated.

The docoll solution

The docoll (*doc*ument *col*lation) system solved all the above problems insofar as it was possible to identify all the folders/directories that contain data on each client computer.

The docoll system is made up of three distinct parts, as follows.

1. Synchronisation

Each client desktop and laptop uses cwrsync or rsync to synchronise configured folders/directories to a dedicated directory on the server for each client, archiving changed files.
 
Synchronisation runs at startup/boot or as otherwise scheduled, typically at least once a day.  The synchronised files are available from a CIFS file system on the server, allowing user-driven restores in the event of loss.  CIFS was formerly SMB, also known as "Windows share".

2. Collating and indexing

Running on the server, scripts periodically collate text-searchable files from "sources", including the dedicated directory for each desktop and laptop, into the document collation.  Files with identical content are identified and only one is added to the collation but all paths are added to the collation; if they have diffrent modification times, the earliest is applied to the collation's copy.

A script runs Xapian's omindex to index the collation using configurable commands (filters) to convert files to text.  If several filters are configured for a file type, the script chooses one at random, thus eventually applying all filters to any files that resist conversion (no filters are 100% effective).  

3. Interactive searching

Xapian's omega is used to provide interactive searching of the collation via a web browser.  docoll includes custom templates for omega to build the HTML from.

Multiple instances

Multiple instances of docoll can be run on a server.  These can be used for different organisations, departments etc.

How to get started

Read "docoll system introduction".  Get it by clicking on "Download" -> latest version -> Documentation.  When you have read enough, download the software and install or, in case of doubt, mail questions to -email is unavailable-

Registration Date: Wed 19 Oct 2011 08:21:32 AM UTC
License: GNU General Public License v2 or later
Development Status: 5 - Production/Stable

 

Latest News rss feed
Version 0.7.6.1 released
     posted by catkin, Sat 26 May 2012 06:14:48 AM UTC

An inter-release bugfix release, partly replacing 0.7.6.  Only the Windows rsysnc client's docoll_cwrsyc.vbs script is changed.

Version: 0.7.6.1, 26 May 2012 Bugfix release

  • Windows rsync client synchronisation script:
    • Fixed bug 35653 "The "_Changed and deleted files" contents on the server go ever deeper"
Version 0.7.6 released
     posted by catkin, Mon 02 Apr 2012 12:11:26 PM UTC

Version: 0.7.6, 2 April 2012

  • GNU/Linux rsync client: improved PID file lock file handling.
  • Windows rsync client synchronisation script:
    • Fixed bug 35650 "Log file name components lack leading zeroes"
    • Fixed bug 35634 "docoll_cwrsync.ini: line with single space causes error"
[...]
Version 0.7.5 released
     posted by catkin, Thu 09 Feb 2012 11:29:37 AM UTC

Version: 0.7.5, 9 February 2012

  • GNU/Linux rsync client: no changes.
  • Windows rsync client:
    • Windows 7 64-bit support.
    • Moved docoll files to docoll sub-folder of cwRsync folder.
    • Introduced conf, logs and samples sub-folders.
    • Commented out sample lines in sample C_files_from.txt.
  • rsync server: no changes.
  • Collating and indexing: no changes.
  • Interactive search: no changes. 
  • Documentation: minor improvements.
Version 0.7.4 released
     posted by catkin, Mon 23 Jan 2012 06:48:06 AM UTC

Version: 0.7.4 released 23 January 2012

  • Windows client synchronisation script: 
    • Re-written in VBscript. 
    • Logging introduced.
    • Does not require UnixUtils date.exe.
    • Configuration files:  single .ini file, replacing several .txt files.
[...]
Quick Overview
 Memberlist (1 member)

Communication Tools
 Mailing Lists (1 public mailing list)

Development Tools
 Bug Tracker (open items: 5, total: 10)
 Task Manager (open items: 0, total: 0)

Back to the top

Powered by Savane 3.13-758e.
Corresponding source code