Code Sprint Tomorrow

[Audrey Eschright]

Sorry about the late notice to the mailing list on this--the core team has been very busy with work and possibly a little forgetful.

We're having a code sprint (Halloween edition!) tomorrow at NedSpace Old Town, from 10AM until 4 or 5 pm. We'll be reviewing in-progress code and patches, looking through the ticket tracker, and generally working on making Calagator better. Anyone and everyone is welcome to come help and contribute, whether a developer, designer, or other interested user. We'll do our best to put you to work.

http://calagator.org/events/1250459316

Audrey

[Igal Koshevoy]

My suggestion for this code sprint: finish the good work we started before. :D

We have a number of features and fixes that we've started work on but
ran out time. Some are probably done and just need a final code
review. Others need quite a bit more development. Many are in between.
I've taken a shot at identifying the branches, their state, described
them, noted any cautionary issues, and described the sort of help that
we need to complete them. I'd like to see us complete a couple of
these at the sprint.

Each of the features and fixes are represented by git branches. The
lines below that look like "* igal/with_fixed_searching" are git
branches available at URLs like
"http://github.com/igal/calagator/tree/with_fixed_searching":

* igal/with_fixed_searching
STATE: Nearly done?

SUMMARY: Fixes how searching is done to increase the number of
correct matches.

DETAILS: The problem was that the Lucene queries were
insufficiently broad to match some text, some of the #to_solr methods
allowed illegal characters that prevented nearby words from being
searched, and the Lucene fuzzy search prevents some exact matches.

CAUTION: The fix is very useful, but has a major drawback: It
requires the user to reindex their search index to make full use of
the improved search criteria, which takes a while.

HELP
1. Review Ruby code for correctness.
2. Review Lucene query for correctness against manual.
3. Ensure that search works adequately well with old index, so
that folks that forget to reindex don't lose any search functionality.

* igal/with_downloader
STATUS: Probably done

SUMMARY: Improves `rake data:fetch` to display download progress
if `curl` or `wget` are installed.

DETAILS: The `rake data:fetch` task takes minutes to download all
the data off the production http://calagator.org/ website. When this
is done with the default `open-uri` library, it seems like the program
is doing nothing for minutes and is hung. The download progress makes
it obvious that a download is in progress and when to expect it to be
completed.

HELP
1. Review Ruby code for correctness.
2. Manually verify with `curl`, `wget` and `open-uri`.

* igal/with_tag_cloud
STATUS: Probably done

SUMMARY: Adds a tag cloud showing the most popular event tags on
the homepage.

DETAILS: The tag cloud makes it easy to see the sorts of events
happening at a glance on Calagator and find new ones.

HELP
1. Review Ruby code for correctness.

* igal/with_refactoring_after_machine_tags
STATUS: Probably done.

SUMMARY: Improves code added in the machine tags sprint.

DETAILS: Many useful refactorings and tests.

CAUTION: Some of this work may have already been done at
reidab/with_tag_search_sorting -- carefully compare these to avoid
collisions.

HELP
1. Review Ruby code for correctness.
2. Ensure that the refactoring and tests are complete.

* aeschright/rical_remodel
STATUS: More development and testing are needed.

SUMMARY: Incomplete attempt to replace the problematic VPIM iCal
library with better RiCal library.

DETAILS: This is useful and important work that which will
eliminate a lot of insanity and parsing hacks we're currently using to
cope with the horror of VPIM and its ilk. Unfortunately, this branch
needs more work and tests to complete.

CAUTION: This code isn't done and radically changes mission
critical functionality.

HELP
1. Review Ruby code and tests to determine what's left.
2. Complete it.

* reidab/with_tag_search_sorting
STATUS: Nearly done?

SUMMARY: Improves machine tags, and filtering and sorting by tags.

DETAILS: Adds ways to sort tags, filter by tags, and provides
additional machine tag services. Useful!

CAUTION: Some of this work may have already been merged or done at
igal/with_refactoring_after_machine_tags -- carefully compare these to
avoid collisions.

HELP
1. Review code against master and
igal/with_refactoring_after_machine_tags branches to identify any code
that's already implemented, fixes that already have been made, etc.
2. Review and test the code.

-igal