Project Detail

Description

1. The old scripts at my forum do not support message ids beyond 9999, and code patches that were recently installed to allow message ids to increment beyond this 9999 barrier introduced minor flaws which can corrupt the message index under certain conditions (which can be reproduced and demonstrated at my forum testbed). Basically, the patches cannot handle replies to replies to replies of topics, which have message ids in the format of ####-#-#-#. If these issues can be resolved, then the existing patches should work, and new patches will not need to be written. If no additional bugs are discovered in the patches, then this task shouldn't be too challenging.

2. Once the above is accomplished, I have upwards of 40 thousand topics (many with duplicate IDs due to past roll over events) that need to be assigned unique, sequential message IDs so that they may be made accessible again. I have a script that nearly accomplishes this (which also rebuilds corresponding message archives and message control file), but it still requires a little work in order to process the message folders (named in yymmdd format) in the correct order. This script also needs to be extended slightly to rebuild corresponding search files based on the new, unique message ids. Given the availability of the nearly completed script, this task shouldn't be too difficult.

3. Thirdly, the paging method needs to be updated. Right now, a form of paging is achieved in displaying topics by date posted, but through a separate 'message archive' function. A new comprehensive system of pagination needs to be integrated with the main forum index and to display an arbitrary number of topics per page. The requirements are that the new pagination scheme be compatible with existing, index based moderator functions for hiding/archiving messages. This can be achieved by building new, sequentially numbered index.txt files (perhaps with each containing 100 topics, though I'm open to suggestion if a better alternative is discovered) from the message archives. After 100 new topics are posted, a new numbered index.txt file would then be added. This would replace the current scheme, which relies on a single, giant index.txt file (of which there are two copies), and as a consequence it grows to large file sizes that invite data corruption. Because of this, this file has been regularly pruned over the years, which is why the message archives will have to be relied upon for building new index.txt files, as it's the only complete source of data. This task may be somewhat challenging.

4. After the above is accomplished, a new, minor function is needed to tally and display the number of topics, and total articles (including replies posted) via template fields. The search facility also needs some minor work to properly enumerate query results spanning multiple pages. These tasks shouldn't take more than a few minutes to complete.

In summary, the basic tasks that need to be accomplished are to debug patches for allowing messages ids to increment beyond 9999, reassign unique and sequential ids to all message data, implement a new method for paging, create a new function to tally and display the total topics and articles posted, and to correct the display of search results.

FYI, the forum software uses a combination of custom written, and unconventional, Shell and Perl scripts that were authored in 2002. Message and user data is stored in flat text files as opposed to an actual relational database. However, an online forum testbed will be made available, and I will actively work alongside the service provider, answer questions, and point out relevant directory locations, scripts and templates at every step of the way. Any project is a collaboration between service buyer and provider in which creative ideas are exchanged, so real-time communication and feedback over YIM is appreciated. It's not mandatory that the service provider have broadband internet access, but it is preferred.

The server is running FreeBSD and ssh access will be provided, but first your ip address will have to be added to the ssh firewall access list. If you are confident in your Perl/CGI/Shell scripting skills, the tasks outlined in this project should not pose an overwhelming challenge for you. If all goes well, your services may be retained, as several projects are planned for the immediate future due to the age and state of these old scripts.

Please note, I work best with friendly, team-oriented individuals possessing sound English language/communication skills, good work ethic, laid-back attitude and liberal mindset. Priority consideration will be given to those who can begin work soon and respond promptly to private messages. Should your bid be selected, and you do not accept the offer within the hour, the project will be reopened to bidding again. Thanks in advance for your understanding and cooperation!