1 Some Privoxy-related tasks, sorted by the time they
2 have been added, not by priority.
4 The latest version should be available at:
5 https://www.privoxy.org/gitweb/?p=privoxy.git;a=blob_plain;f=TODO;hb=HEAD
7 There's work in progress to fund development on these items using
8 donations. If you want to donate, please have a look at:
9 https://www.privoxy.org/faq/general.html#DONATE
11 1) Add more regression tests. Filters should be tested automatically
12 (variables too). Could probably reuse large parts of Privoxy-Filter-Test.
13 Note that there is currently work in progress to leverage curl's
14 test suite, patches have been submitted upstream:
15 http://curl.haxx.se/mail/lib-2014-06/0070.html
17 3) Fix some more XXX: comments.
19 6) Remove actions that aren't needed anymore:
21 content-type-overwrite should probably stay as it's also
22 used by some of the CGI pages (XXX: name them).
24 crunch-client-header and crunch-server-header should probably
25 go, their only advantage is that their search strings can be
26 controlled through the CGI pages, other than that they only
29 crunch-if-none-match can be replaced with a header filter.
31 prevent-compression has a misleading name and could
32 be replaced with a header filter.
34 7) force-text-mode has a stupid name and should probably
35 be renamed to force-filter-mode.
37 8) handle-as-empty-document and handle-as-image should
38 be merged to something like handle-as{something} to
39 prevent them from being activated at the same time.
41 10) There's a bug in the CGI editor that turns the
42 first section's "Insert new section below" into
43 a "Insert new section above" button.
45 11) CGI templates should use semantically-correct HTML
50 12) Support pipelining for outgoing connections.
52 14) Allow to filter POST parameters.
54 19) enable-forward-fallback. Syntax? Suggested by K.R.
56 21) User Manual delivery doesn't accept multiple slashes. Should it?
58 22) Verify action files properly (Including arguments) and
59 act accordingly (should probably intercept all requests
60 with a "Invalid option foo detected" CGI page).
62 23) Do the same in case of syntax errors in the configuration file,
63 instead of just exiting or ignoring the problem.
65 25) Handle multiple filters with the same name better. Reject them?
67 26) Let show-url-info detect clearly invalid URLs.
69 27) Make errno logging less thread-unsafe.
70 Verify that it's really an improvement.
72 28) Don't take default ports in case of invalid forwarding ports.
74 31) If a string action foo is disabled csp->action->string[ACTION_STRING_FOO]
75 doesn't necessarily contain NULL, but may contain the string of an
76 enabled foo action in an overruled section. Is it a bug? Does it matter?
78 32) In case of forwarding failures with socks port == 9050,
79 show extra info about Tor (the whole FAQ entry?).
81 36) Unload unused action files directly, even if they are
82 disabled without replacement.
84 38) In the final results, explicitly list disabled multi actions
85 with their parameters. Not as trivial as it sounds.
87 40) When running in daemon mode, Privoxy's working directory is '/'
88 which means it may not have permissions to dump core when necessary.
89 Figure out a way to solve this. Introduce a cwd config option?
91 41) Change documentation framework to one that works cross-platform.
92 Evaluate WML and txt2tags.
94 42) Add a DTrace USDT provider. Now that FreeBSD has userland DTrace
95 support there's no longer any reason not to.
97 43) Write a tool to check URL patterns against URLs in the log file.
98 This could be included in Privoxy-Regression-Test.
100 44) Privoxy-Log-Parser: Consider highlighting "Connection" in:
101 23:13:03.506 283b6100 Header: Replaced: 'Connection: Keep-Alive' with 'Connection: close'
103 50) Investigate possible PCRS template speedup when searching
104 macros with strstr() before compiling pcrs commands.
105 Investigated, needs some restructuring but is probably worth it.
107 51) Make user-manual directive more generic to allow serving the FAQ
108 and files from user-specified directories. Consider changing the
109 port for "same origin policy" issues.
111 53) Find a more reliable hoster. Involves finding out what our
112 requirements are and which SF alternatives fulfil them.
113 It would probably also make sense to look into what other
114 projects did when migrating away from SF.
116 2014-05: Work in progress. Hosting wish list at the end
117 of this file. Looks like most of the other projects
118 that left SF had lower standards and moved to hosters
119 that don't come close to sattisfying the requirements.
120 2016-03: The website has been moved away from SF infrastructure
121 and is also available through https:// now.
122 2016-04: Server rent for a year has been sponsored by ChameleonJohn.
123 2016-04: The SF mailing lists have been deprecated, the new ones
124 are available at: https://lists.privoxy.org/
126 Interested donors: 1.
128 58) Move more template strings from the code into the actual templates.
130 59) Import the German template translation.
132 60) Ask the Russian translators for input on how to make their
135 61) Consider (optionally?) skipping the hostname comparison when
136 checking if a connections that goes to a HTTP proxy can be reused.
137 Do all HTTP proxy support that? Is it worth it?
139 63) Reject clearly too large requests earlier?
141 64) Use proper copyright attribution. "Privoxy Developers"
144 65) Polish Website. Probably involves ditching the Docbook
145 mess. There are already several threads in the mailinglist
146 archives about this. See also #41.
148 66) Stop hard-coding the number of action and filter files.
150 67) Clean up source code directory layout.
152 68) Use standard make syntax so we don't depend on GNU make.
154 69) Update autoconf setup (or move away from it).
155 Unfortunately the autoconf files can't be simply updated
156 due to license issues:
157 https://lists.privoxy.org/pipermail/privoxy-devel/2016-April/000008.html
159 70) If the server connection is reset but the headers are
160 received, consider passing the mess to the client instead
161 of showing the connect-failed template. Relates to #2698674.
163 74) Let Privoxy-Regression-Test optionally check that action
164 sections which disable actions actually are preceded by
165 sections that enable said actions.
167 75) Create a tool that creates Privoxy action (and filter?) files
168 out of adblock files. Could be implemented as option for
169 url-pattern-translator.pl.
171 76) Cache DNS responses. Note that this has been requested
172 several times by users, but is not a developer priority.
173 If you care about this, feel free to submit patches.
175 77) Allow to configure the IP address used in outgoing connections.
177 78) Allow to optionally use pcre's DFA algorithm.
179 79) Evaluate pcre alternatives.
181 82) Detect if the system time goes back in time let the user
182 know if it caused any connections to get closed.
184 85) Once #51 is done, write a script that populates a directory with
185 various common third-party icons (stumbleupon.png, facebook.png ...)
186 and redirect requests for them to Privoxy.
188 86) Add a server-body-tagger action. This is trivial as as all the
189 functionality required to do it already exists.
191 87) Add a client-body-tagger action. This is less trivial as we currently
192 don't buffer client bodies. After 14) is implemented it would be
195 88) Investigate if there's a Perl module that Privoxy-Regression-Test
196 could optionally use to keep connections alive, preferably while
197 requiring less forks at the same time.
199 89) When multiple block actions apply, consider showing all the block
200 reasons on the blocked page that haven't been overruled, not just
203 91) Add an optional limit for internal redirects. It would probably
204 be reasonable to default to a limit of one and showing an error
205 message if the request for the redirect URL would be redirected
208 92) The statistics currently aren't calculated correctly by Privoxy
209 as each thread is only counted as one request which is no longer
210 correct. This should be fixed, or the statistic code removed.
211 Privoxy-Log-Parser's provides more detailed statistics, anyway.
213 93) Add a config directive to let Privoxy explicitly request either
214 IPv4 (or IPv6) addresses, even if the system supports both.
215 Could be useful as a workaround for misconfigured setups where the
216 libc returns IPv6 addresses even if there's no IPv6 connectivity.
218 94) Add a config directive to let Privoxy prefer either IPv4 (or IPv6)
219 addresses, instead of trusting the libc to return them in an order
220 that makes sense. Like #93, this could be useful as a workaround
221 for misconfigured setups.
223 96) Filters should be easier to look up. Currently get_filter() has to
224 go through all filters and skip the filter types the caller isn't
227 98) When showing action sections on the CGI pages, properly escape
228 line breaks so they can be copy&pasted into action files without
231 99) Figure out a mechanism through which a user can easily enable
232 site-specific action sections that are too aggressive to be
233 enabled by default. This could be similar to the presettings
234 in default.action, but could also be just another action file
235 that isn't used by default.
237 100) Create a cross-platform Privoxy control program and retire
238 the win32 GUI. Integrate support for Privoxy-Regression-Test,
239 Privoxy-Log-Parser, Privoxy-Filter-Test, uagen and similar tools.
240 Interested donors: 1.
242 102) Add an include directive to split the config file into several parts.
244 103) Potential performance improvement for large action files:
245 when figuring out which actions apply, check the action bit mask
246 before pattern matching and skip section that wouldn't modify the
247 actions already set. To increase the impact the sections would have
248 to be applied in reverse.
250 104) The code to modify global_toggle_state should be factored out into
251 a separate function. Currently we mess with it in three different
252 files, but only in w32log.c the tray icon is explicitly set.
253 The logging is inconsistent as well. For details see #3525694.
255 106) actionlist.h should be embedded in a way that causes less text
258 107) Support more pcrs variables, for example $destination-ip-address
259 and $source-ip-address.
261 108) Allow to use a somewhat random string instead of PRIVOXY-FORCE.
263 109) Let log_error() support the format specifier %S which should
264 work like %s but escape new lines like %N. This would be useful
265 to log the result of header filters which may inject new lines.
267 110) Add a global-buffer-limit directive that roughly limits how
268 much malloc'ed memory Privoxy will use and can potentially
269 be smaller than (buffer-limit * max-client-connections).
271 111) Reject requests if hosts and ports in request line and Host
272 header don't match (before filters have been applied).
274 112) If a header filter is used to inject another header by inserting
275 a \r\n (undocumented feature), detect it and split the headers so
276 following header actions do not treat them as a single string.
277 Alternatively add another header injection mechanism.
279 113) Log statistics upon receiving a certain signal (SIGINFO or SIGUSR1).
281 114) Properly deal with status code 100. The current "Continue hack"
282 can cause problems for gpg when uploading keys through Privoxy.
284 115) Add ICAP (RFC 3507) support. FR #3615158.
286 116) Due to the use of sscanf(), Privoxy currently will fail to properly
287 parse chunks whose size can't be represented with 32 bit. This is
288 unlikely to cause problems in the real world, but should eventually
289 be fixed anyway. See also:
290 https://bugzilla.mozilla.org/show_bug.cgi?id=959100
292 118) There should be "escaped" dynamic variables that are guaranteed
293 not to break filters.
295 120) Add an option to limit pcre's recursion limit below the default.
296 On some platforms the recursion limit doesn't prevent pcre from
297 running out of stack space, causing the kernel to kill Privoxy
300 121) Add HTTP/2 support. As a first step, incoming HTTP/1.x requests
301 should be translated to outgoing HTTP/2 requests where possible
302 (and if desired by the user).
303 Interested donors: 1.
305 122) Allow customized log messages.
307 124) Add support for the "lightweight OS capability and sandbox framework"
308 Capsicum. http://www.cl.cam.ac.uk/research/security/capsicum/
309 Interested donors: 1.
311 125) Allow clients to HTTPS-encrypt the proxy connection.
312 Interested donors: 1.
314 126) Run the Co-Advisor HTTP compliance tests, evaluate the results,
315 fix the compliance issues that aren't by design and document
317 Note that Privoxy developers qualify for free account upgrades:
318 http://coad.measurement-factory.com/details.html#pricing
320 127) Add "real" CGI support (serve program output instead of forwarding
321 the request). The work is mostly done due to +external-filter{}.
323 128) Add a config directive to control the stack limit.
325 129) Completely implement RFC 7230 4.1 (Chunked Transfer Coding).
326 Currently Privoxy doesn't properly deal with trailers which
327 are rarely used in the real world but should be supported anyway.
329 130) Move header_tagger() out of the parser structs and let it execute
330 taggers one-by-one against all headers so the header order has less
331 influence on the tagging result. As a bonus, dynamic taggers would
332 have to be compiled less often.
334 131) The handle-as-empty-doc-returns-ok directive should be replaced with
335 an action so the behaviour can be enabled on a per-request basis.
336 Interested donors: 1.
338 133) Consider allowing bitcoin donations.
339 Interested donors: 2.
341 134) Track the total number of bytes written to and received from a socket.
343 135) Add OpenBSM audit support.
345 136) Make builds reproducible.
347 137) Add a (preferably vector-based) logo.
349 138) Bring back the scripts to provide actions file feedback.
351 Once upon a time (~2003) there were scripts on the webserver
352 to make reporting action file feedback more convenient for the
353 user and the actual reports more useful for the developers.
354 They have been unusable for years and have thus been disabled,
355 but making the reporting mechanism available again would be a
358 140) Toggling Privoxy off currently also disables stuff that
359 probably shouldn't be affected (such as actions like
360 forward-override). Investigate and fix or document.
362 142) Remove or update the "internal" pcre version.
364 143) Add support for OpenBSD's pledge feature once it's stablelized.
365 This should be a lot less work then #124.
367 146) Allow to save the internal client tag state to disk and
368 load it after restarts.
370 147) Improve "Building from Source" section in the user manual.
371 A common problem seems to be that it's not obvious to non-technical
372 users how the listed dependencies can be installed on the commonly
373 used platforms. Adding a couple of examples should also be useful for
374 technical users (like Privoxy developers) who want to install or test
375 Privoxy on platforms they are not familiar with.
377 148) Add a config directive to change the CGI_SITE_2_HOST
378 (default: config.privoxy.org).
380 If Privoxy is used as reverse proxy or intercepting proxy without
381 getting intercepted requests, error pages created from default templates
382 currently can result in client requests to config.privoxy.org on the
383 Internet which may not be desirable.
385 150) Add blacklistd support.
387 151) Let the dok-tidy target work cross-platform without introducing
388 a ton of white-space changes that hide the content changes.
390 152) Fix CSS references in the website documentation.
391 For many pages p_doc.css is specified twice using different paths.
392 Usually at least one works, but not all of them do and the
393 duplicated requests are pointless even if they don't end up with
396 153) Catch SIGINT and use it to close the listen socket, serve
397 remaining connections and shut down. This would allow higher
398 uptime and make testing more convenient.
400 154) Underline links in docs and cgi pages. More precisely,
401 don't mess with the browser defaults for link underlining.
403 155) The sig_handler() shouldn't call log_error().
404 While it isn't known to cause actual problems in normal operation,
405 it's technically incorrect and causes crashes when running in
408 156) Reject socks requests with an explicit error message similar
409 to the one used for ftp. Motivation:
410 https://lists.privoxy.org/pipermail/privoxy-users/2017-March/000195.html
412 158) Use a single thread to wait for new requests on reused client connections.
413 Currently the thread that handles the first request on a connection
414 stays responsible for the client connect until it gets closed.
415 In case of lots of idle connections lots of waiting threads are used.
416 While it's conceivable that this ineffiency is irrelevant from a
417 performance point of view, using a single thread should reduce Privoxy's
418 memory footprint a bit which may be noticeable in case of multi-user setups
419 with hundreds of idle connections.
421 161) Properly support requests with chunked transfer-encoding with https inspection.
423 162) When https inspecting, delete generated keys and certificates if
424 the connection to the destination could not be established.
425 Makes silly DoS attacks slightly more complicated.
427 163) Use subdirectories in the certificate-directory to lower the number
428 of files per directory.
430 164) Evaluate switching from pcreposix(3) to pcre's native api
431 for URL matching which allows to compile the patterns once
434 165) Add a max-connections-per-client directive.
436 166) Figure out how to ship Windows binaries with external libraries
437 like pcre and MbedTLS. Required for #142. Somewhat related:
438 https://lists.privoxy.org/pipermail/privoxy-devel/2020-November/000400.html
440 167) Set up a public Privoxy-Filter-Test instance.
442 168) Add a privacy policy.
444 ##########################################################################
446 Hosting wish list (relevant for #53)
451 - Mailinglists (Mailman with public archives preferred)
452 - Webspace (on a Unix-like OS that works with the webserver targets
455 - Commit mails (preferably with unified diffs)
457 (Unsorted) details to look at when evaluating hosters:
459 1. Preferably no third-party ads and trackers.
460 External images, CSS and JavaScript may count as trackers
461 but texts like "supported by company XYZ" may be acceptable.
463 2. JavaScript should be optional or not used at all.
465 3. Services we don't need shouldn't be enabled anyway.
466 (We currently don't use Web forums, wikis, surveys etc.)
468 4. It would be preferable if the hoster didn't have a bad track
469 record as far as user experience, security and privacy are
470 concerned and if the terms of service are "reasonable" and
471 haven't changed too often in the past. Updates in the past
472 should have been improvements and not regressions.
474 5. It would be preferable if most of the server administration
475 is done by a trusted third-party (or at least not a lot of work
478 6. The server(s) should be located in a country with laws we can
479 understand and follow (or at least not unintentionally violate).
481 7. A server location in a country with some kind of due process
482 and strong data protection laws (at least on paper) would be
485 8. Given that Privoxy is a free software project it would be
486 preferable if the hoster would use free software where possible.
488 9. Migrating away from the hoster in the future without losing
489 any important data should be possible without writing web