[
https://issues.apache.org/jira/browse/WHIMSY-4?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=14716699#comment-14716699
]
Sam Ruby commented on WHIMSY-4:
-------------------------------
bq. I assume that URL is the final one?
Unless I hear any input requesting a change, then sure. Note that this data
isn't being updated just yet, I'm working through how the changes to the
underlying libraries affect other tools before deploying.
bq. I suppose this could either be done by adding another date to the file, or
by ensuring that the cache file is touched (so the reader can use the HTTP date)
Defeating HTTP caching will slow things down. I'd like to understand the use
case further before proceeding. I'll also note that whimsy has a cron job that
does a single SVN update of the underlying data for all of the various tools to
use, and the script to produce the public committee-info.txt would be a
separate cron job. Which means that if the first cron job were to ever fail,
the second would continue to report that it had run. My take is that if a
human ever suspects that the data is stale, the current information is all they
need to confirm it.
Can you identify an action an automated tool would take with access to this
data without user interaction?
bq. The "report" attribute seems incomplete: if the cttee is on a monthly
cycle, then the default cycle is not shown at all. I think it would be better
to show both sets of data. Also, rather than a single string containing
comma-space separated entries, it might be easier to use if the months were
separate entries in an array. i.e. one entry for each month, plus one starting
"Next month:" (if needed)
Sounds reasonable, but will need to be implemented in a way that either doesn't
affect existing tools or those tools need to be changed. Shouldn't be
difficult, but I would like to deploy what I have before I make this change.
> Create public sanitised version of committee-info.txt
> -----------------------------------------------------
>
> Key: WHIMSY-4
> URL: https://issues.apache.org/jira/browse/WHIMSY-4
> Project: Whimsy
> Issue Type: New Feature
> Reporter: Sebb
>
> The committee-info.txt (CI) file contains the canonical information for the
> composition of PMCs etc. This information is needed by various other scripts
> and is supposed to be synched to various web pages.
> It would be very useful to create a sanitised, structured version of the CI
> file for consumption by other apps, for example people.apache.org,
> projects.apache.org, reporter.apache.org.
> Whimsy already has access to the file, it would just need to reformat it and
> publish it on the public part of the site.
> The advantage of using Whimsy to create a sanitised version is that only one
> place would be responsible for the cleaning up process.
> Ideally the format would be readable by both humans and computers.
> So perhaps JSON would be suitable (Python supports this).
> Probably best to recreate the file on a timed basis rather than as an
> svnpubsub client. [In the case of svnpubsub it may not be easy to determine
> if the process has stopped working, or there have been no recent updates to
> the file]. In any case the file should contain a timestamp.
> The following data is needed as a minimum:
> PMC names (foundation date) and chairs (name and availid)
> PMC members (name and availid) and joining dates
> Similar info for the other Committees (Board, President's, W3C)
> Reporting schedules:
> - quarterly months with contents
> - next month list (not sure if comments are needed)
> There are some PMCs that were re-established after having been disbanded.
> In this case the dates should probably be taken from the most recent
> incarnation?
--
This message was sent by Atlassian JIRA
(v6.3.4#6332)
