| Age | Commit message (Collapse) | Author |
|---|
|
Corrects pronunction by screen readers and improves semantics.
Reviewed-by: Eliot Whalan <ewhal@pantsu.cat>
|
|
Fixes a bug where Response class `error()` function may blow up and
invalid HTTP status code `42` if `PDO::ATTR_ERRMODE` is set to
`PDO::ERRMODE_EXCEPTION` during PDO initialization or query.
Clarifies what HTTP status codes 200 and 500 are for readability.
|
|
Add a comment explaining how HTTP status code 200 means success.
Cherry-picked from master 4606553f38909b483bc06674f9558ede1a9ff1f6.
|
|
Fix a bug where Response class `error()` function may blow up and return
an invalid HTTP status code `42` if `PDO::ATTR_ERRMODE` is set to
`PDO::ERRMODE_EXCEPTION` during PDO initialization or query. By default,
it is implicitly set to `PDO::ERRMODE_SILENT`.
This bug can occur when a MySQL server is running, but encounters an
error post-login (e.g. table doesn't exist).
|
|
These commits made some recent changes to the FAQ in year 2016:
- FAQ: Change non-abuse contact to info@pantsu.cat
- FAQ: Avoid using word "content" as a noun
These may be covered by copyright.
|
|
According to the GNU Project, using the noun "content" to describe
publications and works of authorship "adopts an attitude you might
rather avoid: it treats them as a commodity whose purpose is to fill a
box and make money. In effect, it disparages the works themselves."
From now on, we'll refer to publications as uploaded or published works.
This bug was introduced in commit
7d61e0439893a2fd946ec4d4b8e32aaaac0da322.
See: https://www.gnu.org/philosophy/words-to-avoid.en.html#Content
|
|
This marks the Pomf 2.0.2 release.
Reviewed-by: Eliot Whalan <ewhal@pantsu.cat>
|
|
Bootstrap is a CSS framework by Twitter, Inc.
Pomf uses certain Bootstrap 3 elements, such as .jumbotron and .alert.
The credit is due for Twitter, Inc. for Bootstrap, which like Pomf is
free software licensed under the Expat license.
The CSS snippets about Bootstrap were initially introduced to
style.min.css in commit df6cf936d7ef3a819c6ff484b24a8cb5946726a9, then
moved into pomf.css uncompressed in commit
152c0257a8ef9a8da7e45717a65578c340a7c178.
See: https://getbootstrap.com/
Reviewed-by: Eliot Whalan <ewhal@pantsu.cat>
|
|
It turns out that moving to Rizon's IRC network was a bad idea after
all. One or more developers (myself included) are unable to connect to
Rizon for using Tor or operating Tor exits.
Now there's this split: Users and Pomf site maintainers are on #pomfret
on Rizon (where I don't have frequent access) and development has moved
back to #pomf on Partyvan (where there's also #/g/hostingtalk).
Given how I personally play such a big role in leading Pomf development,
it has no longer been practical to continue development communication on
an IRC network I can't access. There was no interest to finish setting
up a mailing list for Pomf development because IRC was considered more
important.
Thus, the stagnation and stalled development of pantsu/pomf between
2016-01-31 and 2016-04-30 as seen in commit logs.
There used to be an IRC channel called #pantsucat on Partyvan's IRC
network, before development moved to Rizon. That channel is no longer
active.
See: pantsu/pomf@5846ba7ccb1aa6af2850b3fcf34c6bd396995161
Reviewed-by: Eliot Whalan <ewhal@pantsu.cat>
|
|
Code cleanup, that's all.
Reviewed-by: Eliot Whalan <ewhal@pantsu.cat>
|
|
One of NPM's rules [1] is: "New packages must not have uppercase letters
in the name." This also follows the convention where Git repository
names are lowercase.
This change is mostly irrelevant to us, but it's a bug nonetheless.
There's no concern over backwards compatibility regarding SemVer as far
as I'm aware.
This bug was introduced in 7d28ef53a8166cbdcb558a84d5a53cfd73472438.
[1]: https://docs.npmjs.com/files/package.json
Reviewed-by: Eliot Whalan <ewhal@pantsu.cat>
|
|
Actually on second thought, SemVer doesn't allow us to remove
functionality without incrementing the major version number... which we
don't want to do for this.
Set the bugs email address to bugs@pantsu.cat. Hoping here bugs@ to be
an alias to ewhal@ for the time being until a real mailing list will be
reachable at that address. This fixes the issue for the time being
without removing functionality.
It is okay that we made a mistake. We don't need to increment the major
version here now, because no release was tagged and released in this
branch.
Nevermind that Pomf isn't even published on NPM packages, so this
"feature" is never seen in its current form in the npm(1) program. We'll
still respect SemVer, for what little worth it may be to other tools.
This reverts commit 310d8d9fdd33826b9a6d5c1f9b1f8aa1c373c461.
Reviewed-by: Eliot Whalan <ewhal@pantsu.cat>
|
|
Why is this a thing? We never used pantsu/pomf issue tracking. It was
disabled a long time ago in gogs, and now it doesn't even exist.
For consideration, we could provide an email address here. There's no
mailing list for bugs (yet) and it's unlikely to see one soon.
Compliments commit bf4fbf5f8718c4668a57d7dc09880deb904fef1b. Inspired by
~alucard/pomf at Cuntflaps, though never reported to us.
Reviewed-by: Eliot Whalan <ewhal@pantsu.cat>
|
|
git.pantsu.cat has switched from gogs to cgit. gogs used a slightly
different URLs to point to tree, so the old URLs from gogs are no longer
valid.
For consideration, to ease maintenance we'll just keep pointing the URL
to the master branch and not a specific commit or maint branch.
?h=master is not really required here, it's the default branch. I am
aware that the current maint and master branches diverge heavily from
each other in terms of JavaScript, but this will be an issue that will
disappear eventually. Thus, this is the simplest fix.
A HTTP redirect will be setup for the old URL to the new one
(eventually) to fix broken links.
Reviewed-by: Eliot Whalan <ewhal@pantsu.cat>
|
|
This marks the Pomf 2.0.1 release.
Sorry again that the changelogs didn't make it to this release - we're
working on it. Hopefully.
|
|
- Add missing newline to Response.class.php
- Fix undeclared vars in html_success() and text_success()
Reviewed-by: Eliot Whalan <ewhal@pantsu.cat>
Reviewed-by: clsr <clsr@clsr.net>
|
|
As explained by clsr:
> <clsr> With default PHP settings, that'd print a warning, then treat
> $result as if it was an empty string before appending.
This bug was introduced in commit
a92961f24072bc4c3881bf0e2f609562f3703471 and
1411bce1dc13bf6fea66b5899ba29d50d5c53bb0.
Reported-by: clsr <clsr@clsr.net>
Suggested-by: clsr <clsr@clsr.net>
|
|
Right before text_error(), there was a missing newline. This bug was
introduced in commit a92961f24072bc4c3881bf0e2f609562f3703471.
|
|
Remove page specific style rules. These are unused relics from before
2.0.0 release.
This merge compliments 7743e0ec4e1c71e7cc0e0df6203e02d221c82721, which
was merged too early.
|
|
The tools page has only had alive and maintained tools on it since Pomf
2.0.0. Ded and old tools have been removed in commit
627937d0fa632a677f278af467886915c0a36f35.
It is anticipated that these style rules will be never used again. If we
did, we can add them back later: "you aren't gonna need it" (YAGNI). If
it's dead, you should probably remove the tool from the tools page
instead of beating a dead horse.
[wub@partyvan.eu: Modified Git commit message, as discussed]
Signed-off-by: Juuso Lapinlampi <wub@partyvan.eu>
|
|
Remove inclusion of a dead template in JavaScript-free uploader.
Reviewed-by: Eliot Whalan <ewhal@pantsu.cat>
|
|
The JavaScript-free uploader doesn't care about browser features too
much, specifically about window.FormData. That API is for JavaScript.
Because there also isn't any JavaScript to execute, this template never
gets called. It's dead HTML, so remove it.
|
|
Fix some 404 and outdated links in the tools page.
- Update ShareX URL and settings
- Update Gyazowin source code link
- Rename Gyazo Client to Gyazowin for simplicity
Reviewed-by: Eliot Whalan <ewhal@pantsu.cat>
|
|
Acked-by: Eliot Whalan <ewhal@pantsu.cat>
|
|
This outline-offset value has been commented out for how long, and it
seems to have no purpose. Remove it.
Signed-off-by: Juuso Lapinlampi <wub@partyvan.eu>
|
|
"Gyazo Client (Puush like client)" is a horrible word monster. In the
Pomf community, it is better known as Gyazowin which refers to the
actual project and repository name.
Call it Gyazowin from now on in the tools page.
|
|
Pantsu.cat's fork of Gyazowin (aka "Gyazo Client (Puush like client)")
had moved its development to git.pantsu.cat somewhere around September
23, 2015. This link to GitHub is now dead (404 Not Found).
Replace it with the repository hosted on git.pantsu.cat, at least until
ewhal@ decides to get rid of that unmaintained repository.
Reported-by: Jaex
|
|
ShareX has moved its source code repository to GitHub. Google Code is
also shutting down, so update and point the URL to the GitHub
repository.
While we are here, update and sanitize the ShareX settings we need to
use. Also update the copyright year in header.
Reported-by: Jaex
Suggested-by: Jaex
|
|
We are ready to release Pomf 2.0.0. Patches will follow semantic
versioning, so th next 2.0 branch release will be 2.0.1. The next
feature release will be the 2.1 branch with release 2.1.0.
OK ewhal@
|
|
New code has been added to Response.class.php this year in commit
b94ac9a. Update LICENSE and Response.class.php copyright header with
year 2016.
We can't do this for every file in the project repository unless they
are actually changed, due to the way our copyright legislation works. We
also can't remove the mention of 2015, because most of the code was
initially published in 2015 and so remains copyrighted since that year.
|
|
Clean up a lot of the cruft in Response class.
Gyazo is deprecated and falls back to 'text' for similar functionality,
reducing code duplication. It will be removed in a future release.
CSV gets a text/csv Content-Type. CSV linebreaks are also now CRLF, and
values are wrapped in quotes.
The default error is now a JSON response instead of a generic 400 Bad
Request in plaintext.
output=json is now explicit and no longer undefined behavior.
|
|
The gyazo response would return an empty response, because the construct
didn't actually fallthrough to text properly. Force gyazo's response
type to be 'text' to fix the issue.
Accordingly, the 'gyazo' cases for send() and error() now become dead
code. Remove them.
OK ewhal@
|
|
Add myself to package.json contributors. I have a significant amount of
code and effort dedicated to this project, so I can be contacted for
questions.
OK ewhal@
|
|
Lint our spaghetti JavaScript code to be at least slightly more
readable. Configuration used is ESLint 1.7.x and airbnb/legacy.
It's been so long I cannot remember every change that was done, but long
story short: all the ESLint errors are now gone, at least at the date of
last commit.
moep contributed some ideas for patches: some were implemented, some
were discarded as the time and patience for merging this ran out. Thus,
there are still ESLint warnings about anonymous functions.
I don't feel like spending much more time on this. The future Pomf 2.1
development branch will replace the JavaScript code in its entirety,
courtesy of luminarys@.
Suggested-by: moep
Tested-by: Eliot Whalan <ewhal@pantsu.cat>
OK ewhal@, maybe a vague OK from luminarys@ a long time ago
|
|
|
|
Reword the "no Git issue tracking" policy in contributing section.
Previously the contributing section in README.md would say we don't use
git.pantsu.cat's issue tracker to track issues. That Git service
repository's issues have been disabled today, so we might as well say we
don't use any Git service issue tracking. Unofficial mirrors like the
one on notabug.org may still have issue tracking, which I believe we
have no control over.
This commit also clarifies that we do it for ethical and technical
objections. (In short, it's because the Git service is centralized and a
walled garden.) For more information, see
wubthecaptain/deathwatch-pomf@fc86cbf for information why.
|
|
Peter Lejeck's (peter.lejeck@) only commit in the file is a507748, which
is pretty much just the HTML layout in the file. I have changed that, a
lot.
Eric Johansson's (neku@) commits in the file covered the text itself.
This too has been replaced in commit 7d61e04, which rewrote all the
text in the page.
In other words, these two authors no longer have any meaningful content
in the file to be attributed. Much of the copyright goes to the
Pantsu.cat developers - which covers myself too.
OK luminarys@, "no complaints"
|
|
Fixup to previous commit 120627d. Fixes Grunt build failure.
Gladly, I am working on replacing Grunt with Makefile for 2.1.0 or 3.0.0
release so that these mistakes should happen less frequently.
|
|
Catalan translation was removed in commit 566611c. However, the commit
was incomplete and forgot to remove the swig build task from Grunt.
We can safely remove these Catalan pages from Grunt swig build. They no
longer exist and this is dead code.
OK luminarys@
|
|
Previous commit c950e85 came with a workaround to resolve a TypeError in
EventEmitter.prototype.emit for-loop. That workaround sacrificed some
readability for bug fixing due to technical debt.
bursey came up with a better solution on IRC which seems to work and
improves readability. Something needs to emit an event that wasn't bound
with "on", thus the event was not in the _events object.
Airbnb ES5 style guide requires to have all variables at top, so we
leave the len variable undefined at the top to be defined later. After
the _events check, we define len to avoid the TypeError issue while
conforming to Airbnb ES5 style guide and improving readability.
This commit also adds comments to document the function.
Suggested-by: bursey (#/g/partyvan @ Partyvan IRC)
|
|
this._events[evt].length would run into JavaScript scope issues in
EventEmitter.prototype.emit function for-loop.
This is an attempt to work around the issue to resolve an innocent
looking TypeError. The code seems to work without this patch, but I
can't spot if this is some dead code or know what to do with it. Due to
technical debt, some sacrifice had to be made in readability for fixing
this bug.
This bug was introduced in commit 575b78b.
For future improvement, this function and other functions in the file
should be refactored or replaced with a more efficient implementation.
|
|
Very short identifier names usually make the code harder to read and
potentially less maintainable.
Fixes two ESLint id-length errors using airbnb/legacy config.
Suggested-by: avail <avail@pomf.se>
|
|
- Rephrase check_script.swig JavaScript warning
- Remove check_script URL dependency to no.pantsu.cat
OK ewhal@
|
|
check_script.swig would hardlink to no.pantsu.cat domain for the purpose
of Catalan translations as part of a hack. Catalan translations were
removed in commit 566611c, so we can link to the nojs.html page directly
(built at compile time).
This assures that all sites using this project's code will have less to
tweak to make this project work on their site.
|
|
Clarify that the site requires JavaScript to function and what the user
should do. Telling the users only "Javascript free version" is confusing
at best, and we can do better.
For future improvement, the JavaScript-free version may be integrated on
the same page as the JavaScript uploader.
|
|
When a local variable shares the same name with a variable in its scope,
it can cause confusion to the reader and make it impossible to access
the global variable.
Disallow shadowing. Fixes ESLint no-shadow errors using airbnb/legacy
config.
Suggested-by: marc <mharnos@gmail.com>
|
|
Quoting ESLint:
> Variables that are declared and not used anywhere in the code are most
> likely an error due to incomplete refactoring. Such variables take up
> space in the code and can lead to confusion by readers.
This commit removes them. Fixes ESLint no-unused-vars errors using
airbnb/legacy config.
Suggested-by: marc <mharnos@gmail.com>
|
|
(Partially reverts commit e0bb4e0.)
Actually fix generate_name() COUNT query this time. It still had an
imaginary table "pomf" from e0bb4e0. Change it to "files" like in the
schema.
For code consistency, :filename is reverted back to :name as it was
before commit e0bb4e0. That commit is bogus and doesn't fix the issue
due to skipping the code review process.
The problem with this COUNT query in first place was that it checked
imaginary columns (name) from an imaginary table (pomf) in the database.
This led to the "same filename" check not working at all, and the other
code only relying on checking the same hash (currently SHA-1) for
returning the filename before generate_name().
By luck, generated filenames could collide and another file of the same
generated filename would be returned, discarding the new upload.
I discovered this bug about a month ago myself, but back then there was
no reliable way to reproduce it and no good description of the bug.
This bug was introduced in commit 5b77a4e.
Reported-by: Vhum Vhum <vhumvhum1@gmail.com>
Suggested-by: Luminarys <luminarys@fuwa.se>
OK ewhal@, luminarys@
|
|
database
|
|
Hidden <input> values did not work as I expected for POST requests. In
GET requests, the hidden values are added as query parameters to the
URL. For POST requests, the hidden inputs are simply more form-data to
upload in the POST request (another "file").
Because of this, the HTML output parameter is never passed to upload.php
and so it returns the default response instead (JSON). The goal is to
use the HTML response.
This reverts commit d6dc103c7229cb0e2e177ce0dfc0cfb528025ca9, where the
bug was introduced.
OK ewhal@
|
