aboutsummaryrefslogtreecommitdiffstats
path: root/app/models/foi_attachment.rb
Commit message (Collapse)AuthorAgeLines
* Rails.env should never not be set, I think this is a relic of RAILS_ENVHenare Degan2013-02-20-5/+1
|
* As we're validating filename with validates_presence_of, which doesn't allow ↵Louise Crow2013-01-14-1/+1
| | | | blanks, ensure_filename! should populate a default filename on a blank filename, as well as on nil.
* Only make a tempfile etc if we're actually going to use it. Keep the body ↵Louise Crow2012-12-06-20/+30
| | | | | | | | call outside the Chdir block - sometimes getting the body may involve opening files too. Conflicts: app/models/foi_attachment.rb
* Use the character set of the attachment to encode the string that we're ↵Louise Crow2012-12-05-0/+14
| | | | pulling out of the file before converting it to our default encoding.
* Use Rails.env as a more reliable way of determining environment - it may not ↵Louise Crow2012-11-29-2/+2
| | | | have come from ENV.
* Time out pdftohtml after 30 secondsRobin Houston2012-08-01-8/+15
| | | | | | | | We have a PDF document that appears to send pdftohtml into a loop where it creates millions of tiny PDF files and consumes ever-increasing amounts of memory. (This document: http://www.whatdotheyknow.com/request/119267/response/296719/attach/5/Document%203.pdf)
* File.dirname(__FILE__) is not necessarily absoluteRobin Houston2012-06-03-1/+1
| | | | | | | It’s weird that there’s so much code here that implicitly assumes File.dirname(__FILE__) is an absolute path, because really in general it very much is not! This assumption was invalidated by the recent bundler-awareness changes.
* Further annotation correctionsSeb Bacon2012-05-24-1/+1
|
* Remove trailing whitespace (to make a cleaner forthcoming merge with ↵Seb Bacon2012-05-15-10/+10
| | | | wombleton:feature/440_sparkly_admin_css)
* Raise exception if $RAILS_ENV is unsetRobin Houston2012-01-29-1/+5
| | | | | If $RAILS_ENV is unset, raise an exception rather than just blithely creating a directory called cache/attachments_.
* Use AlaveteliExternalCommand for unrtfRobin Houston2012-01-17-3/+1
|
* Do not hard-code the path to pdftohtmlRobin Houston2012-01-17-3/+1
|
* To get around occasional race conditions, instead of forcing an fsync, retry ↵Seb Bacon2012-01-13-1/+15
| | | | getting the body a few times before failing. May be related to issue #333 and #339
* Changes to attempt to ensure we always get a consistent state for an ↵Seb Bacon2012-01-12-0/+1
| | | | incoming message's attachments. May fix #339.
* Attempt to fix #333 ('attempt' because hard to test!)Seb Bacon2012-01-11-0/+1
|
* Ensure model schema documentation is up to date.Seb Bacon2012-01-09-0/+16
|
* Ensure we regenerate cache if for some reason it's missingSeb Bacon2012-01-09-1/+7
|
* Don't give error when viewing HTML versions. Fixes issue #321Seb Bacon2012-01-08-2/+2
|
* Further improvements to attachment cachingSeb Bacon2011-12-13-17/+16
|
* WIP on caching attachmentsSeb Bacon2011-12-12-0/+322