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