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.
14 3) Fix some more XXX: comments.
16 6) Remove actions that aren't needed anymore:
18 content-type-overwrite should probably stay as it's also
19 used by some of the CGI pages (XXX: name them).
21 crunch-client-header and crunch-server-header should probably
22 go, their only advantage is that their search strings can be
23 controlled through the CGI pages, other than that they only
26 crunch-if-none-match can be replaced with a header filter.
28 prevent-compression has a misleading name and could
29 be replaced with a header filter.
31 7) force-text-mode has a stupid name and should probably
32 be renamed to force-filter-mode.
34 8) handle-as-empty-document and handle-as-image should
35 be merged to something like handle-as{something} to
36 prevent them from being activated at the same time.
38 10) There's a bug in the CGI editor that turns the
39 first section's "Insert new section below" into
40 a "Insert new section above" button.
42 11) CGI templates should use semantically-correct HTML
47 12) Support pipelining for outgoing connections.
49 19) enable-forward-fallback. Syntax? Suggested by K.R.
51 21) User Manual delivery doesn't accept multiple slashes. Should it?
53 22) Verify action files properly (Including arguments) and
54 act accordingly (should probably intercept all requests
55 with a "Invalid option foo detected" CGI page).
57 23) Do the same in case of syntax errors in the configuration file,
58 instead of just exiting or ignoring the problem.
60 25) Handle multiple filters with the same name better. Reject them?
62 26) Let show-url-info detect clearly invalid URLs.
64 27) Make errno logging less thread-unsafe.
65 Verify that it's really an improvement.
67 28) Don't take default ports in case of invalid forwarding ports.
69 31) If a string action foo is disabled csp->action->string[ACTION_STRING_FOO]
70 doesn't necessarily contain NULL, but may contain the string of an
71 enabled foo action in an overruled section. Is it a bug? Does it matter?
73 32) In case of forwarding failures with socks port == 9050,
74 show extra info about Tor (the whole FAQ entry?).
76 36) Unload unused action files directly, even if they are
77 disabled without replacement.
79 38) In the final results, explicitly list disabled multi actions
80 with their parameters. Not as trivial as it sounds.
82 40) When running in daemon mode, Privoxy's working directory is '/'
83 which means it may not have permissions to dump core when necessary.
84 Figure out a way to solve this. Introduce a cwd config option?
86 41) Change documentation framework to one that works cross-platform.
87 Evaluate WML and txt2tags.
89 42) Add a DTrace USDT provider. Now that FreeBSD has userland DTrace
90 support there's no longer any reason not to.
92 43) Write a tool to check URL patterns against URLs in the log file.
93 This could be included in Privoxy-Regression-Test.
95 44) Privoxy-Log-Parser: Consider highlighting "Connection" in:
96 23:13:03.506 283b6100 Header: Replaced: 'Connection: Keep-Alive' with 'Connection: close'
98 50) Investigate possible PCRS template speedup when searching
99 macros with strstr() before compiling pcrs commands.
100 Investigated, needs some restructuring but is probably worth it.
102 51) Make user-manual directive more generic to allow serving the FAQ
103 and files from user-specified directories. Consider changing the
104 port for "same origin policy" issues.
106 53) Find a more reliable hoster. Involves finding out what our
107 requirements are and which SF alternatives fulfil them.
108 It would probably also make sense to look into what other
109 projects did when migrating away from SF.
111 2014-05: Work in progress. Hosting wish list at the end
112 of this file. Looks like most of the other projects
113 that left SF had lower standards and moved to hosters
114 that don't come close to sattisfying the requirements.
115 2016-03: The website has been moved away from SF infrastructure
116 and is also available through https:// now.
117 2016-04: Server rent for a year has been sponsored by ChameleonJohn.
118 2016-04: The SF mailing lists have been deprecated, the new ones
119 are available at: https://lists.privoxy.org/
121 Interested donors: 1.
123 58) Move more template strings from the code into the actual templates.
125 59) Import the German template translation.
127 60) Ask the Russian translators for input on how to make their
130 61) Consider (optionally?) skipping the hostname comparison when
131 checking if a connections that goes to a HTTP proxy can be reused.
132 Do all HTTP proxy support that? Is it worth it?
134 63) Reject clearly too large requests earlier?
136 64) Use proper copyright attribution. "Privoxy Developers"
139 65) Polish Website. Probably involves ditching the Docbook
140 mess. There are already several threads in the mailinglist
141 archives about this. See also #41.
143 66) Stop hard-coding the number of action and filter files.
145 67) Clean up source code directory layout.
147 68) Use standard make syntax so we don't depend on GNU make.
149 69) Update autoconf setup (or move away from it).
150 Unfortunately the autoconf files can't be simply updated
151 due to license issues:
152 https://lists.privoxy.org/pipermail/privoxy-devel/2016-April/000008.html
154 70) If the server connection is reset but the headers are
155 received, consider passing the mess to the client instead
156 of showing the connect-failed template. Relates to #2698674.
158 74) Let Privoxy-Regression-Test optionally check that action
159 sections which disable actions actually are preceded by
160 sections that enable said actions.
162 75) Create a tool that creates Privoxy action (and filter?) files
163 out of adblock files. Could be implemented as option for
164 url-pattern-translator.pl. Before doing that, the already
165 existing solutions should probably be evaluated to see if
166 they do the job already or could be improved.
168 76) Cache DNS responses. Note that this has been requested
169 several times by users, but is not a developer priority.
170 If you care about this, feel free to submit patches.
172 77) Allow to configure the IP address used in outgoing connections.
174 78) Allow to optionally use pcre's DFA algorithm.
176 79) Evaluate pcre alternatives.
178 82) Detect if the system time goes back in time let the user
179 know if it caused any connections to get closed.
181 85) Once #51 is done, write a script that populates a directory with
182 various common third-party icons (stumbleupon.png, facebook.png ...)
183 and redirect requests for them to Privoxy.
185 86) Add a server-body-tagger action. This is trivial as as all the
186 functionality required to do it already exists.
188 87) Add a client-body-tagger action. Work in progress.
190 88) Investigate if there's a Perl module that Privoxy-Regression-Test
191 could optionally use to keep connections alive, preferably while
192 requiring less forks at the same time.
194 89) When multiple block actions apply, consider showing all the block
195 reasons on the blocked page that haven't been overruled, not just
198 91) Add an optional limit for internal redirects. It would probably
199 be reasonable to default to a limit of one and showing an error
200 message if the request for the redirect URL would be redirected
203 92) The statistics currently aren't calculated correctly by Privoxy
204 as each thread is only counted as one request which is no longer
205 correct. This should be fixed, or the statistic code removed.
206 Privoxy-Log-Parser's provides more detailed statistics, anyway.
208 93) Add a config directive to let Privoxy explicitly request either
209 IPv4 (or IPv6) addresses, even if the system supports both.
210 Could be useful as a workaround for misconfigured setups where the
211 libc returns IPv6 addresses even if there's no IPv6 connectivity.
213 94) Add a config directive to let Privoxy prefer either IPv4 (or IPv6)
214 addresses, instead of trusting the libc to return them in an order
215 that makes sense. Like #93, this could be useful as a workaround
216 for misconfigured setups.
218 96) Filters should be easier to look up. Currently get_filter() has to
219 go through all filters and skip the filter types the caller isn't
222 98) When showing action sections on the CGI pages, properly escape
223 line breaks so they can be copy&pasted into action files without
226 99) Figure out a mechanism through which a user can easily enable
227 site-specific action sections that are too aggressive to be
228 enabled by default. This could be similar to the presettings
229 in default.action, but could also be just another action file
230 that isn't used by default.
232 100) Create a cross-platform Privoxy control program and retire
233 the win32 GUI. Integrate support for Privoxy-Regression-Test,
234 Privoxy-Log-Parser, Privoxy-Filter-Test, uagen and similar tools.
235 Interested donors: 1.
237 102) Add an include directive to split the config file into several parts.
239 103) Potential performance improvement for large action files:
240 when figuring out which actions apply, check the action bit mask
241 before pattern matching and skip section that wouldn't modify the
242 actions already set. To increase the impact the sections would have
243 to be applied in reverse.
245 104) The code to modify global_toggle_state should be factored out into
246 a separate function. Currently we mess with it in three different
247 files, but only in w32log.c the tray icon is explicitly set.
248 The logging is inconsistent as well. For details see #3525694.
250 106) actionlist.h should be embedded in a way that causes less text
253 107) Support more pcrs variables, for example $destination-ip-address
254 and $source-ip-address.
256 108) Allow to use a somewhat random string instead of PRIVOXY-FORCE.
258 109) Let log_error() support the format specifier %S which should
259 work like %s but escape new lines like %N. This would be useful
260 to log the result of header filters which may inject new lines.
262 110) Add a global-buffer-limit directive that roughly limits how
263 much malloc'ed memory Privoxy will use and can potentially
264 be smaller than (buffer-limit * max-client-connections).
266 111) Reject requests if hosts and ports in request line and Host
267 header don't match (before filters have been applied).
269 112) If a header filter is used to inject another header by inserting
270 a \r\n (undocumented feature), detect it and split the headers so
271 following header actions do not treat them as a single string.
272 Alternatively add another header injection mechanism.
274 113) Log statistics upon receiving a certain signal (SIGINFO or SIGUSR1).
276 114) Properly deal with status code 100. The current "Continue hack"
277 can cause problems for gpg when uploading keys through Privoxy.
279 115) Add ICAP (RFC 3507) support. FR #3615158.
281 116) Due to the use of sscanf(), Privoxy currently will fail to properly
282 parse chunks whose size can't be represented with 32 bit. This is
283 unlikely to cause problems in the real world, but should eventually
284 be fixed anyway. See also:
285 https://bugzilla.mozilla.org/show_bug.cgi?id=959100
287 118) There should be "escaped" dynamic variables that are guaranteed
288 not to break filters.
290 120) Add an option to limit pcre's recursion limit below the default.
291 On some platforms the recursion limit doesn't prevent pcre from
292 running out of stack space, causing the kernel to kill Privoxy
295 121) Add HTTP/2 support. As a first step, incoming HTTP/1.x requests
296 should be translated to outgoing HTTP/2 requests where possible
297 (and if desired by the user).
298 Interested donors: 1.
300 122) Allow customized log messages.
302 124) Add support for the "lightweight OS capability and sandbox framework"
303 Capsicum. http://www.cl.cam.ac.uk/research/security/capsicum/
304 Interested donors: 1.
306 125) Allow clients to HTTPS-encrypt the proxy connection.
307 Interested donors: 1.
309 126) Run the Co-Advisor HTTP compliance tests, evaluate the results,
310 fix the compliance issues that aren't by design and document
312 Note that Privoxy developers qualify for free account upgrades:
313 http://coad.measurement-factory.com/details.html#pricing
315 127) Add "real" CGI support (serve program output instead of forwarding
316 the request). The work is mostly done due to +external-filter{}.
318 128) Add a config directive to control the stack limit.
320 129) Completely implement RFC 7230 4.1 (Chunked Transfer Coding).
321 Currently Privoxy doesn't properly deal with trailers which
322 are rarely used in the real world but should be supported anyway.
324 130) Move header_tagger() out of the parser structs and let it execute
325 taggers one-by-one against all headers so the header order has less
326 influence on the tagging result. As a bonus, dynamic taggers would
327 have to be compiled less often.
329 131) The handle-as-empty-doc-returns-ok directive should be replaced with
330 an action so the behaviour can be enabled on a per-request basis.
331 Interested donors: 1.
333 133) Consider allowing bitcoin donations.
334 Interested donors: 2.
336 134) Track the total number of bytes written to and received from a socket.
338 135) Add OpenBSM audit support.
340 136) Make builds reproducible.
342 138) Bring back the scripts to provide actions file feedback.
344 Once upon a time (~2003) there were scripts on the webserver
345 to make reporting action file feedback more convenient for the
346 user and the actual reports more useful for the developers.
347 They have been unusable for years and have thus been disabled,
348 but making the reporting mechanism available again would be a
351 140) Toggling Privoxy off currently also disables stuff that
352 probably shouldn't be affected (such as actions like
353 forward-override). Investigate and fix or document.
355 143) Add support for OpenBSD's pledge feature once it's stablelized.
356 This should be a lot less work then #124.
358 146) Allow to save the internal client tag state to disk and
359 load it after restarts.
361 147) Improve "Building from Source" section in the user manual.
362 A common problem seems to be that it's not obvious to non-technical
363 users how the listed dependencies can be installed on the commonly
364 used platforms. Adding a couple of examples should also be useful for
365 technical users (like Privoxy developers) who want to install or test
366 Privoxy on platforms they are not familiar with.
368 148) Add a config directive to change the CGI_SITE_2_HOST
369 (default: config.privoxy.org).
371 If Privoxy is used as reverse proxy or intercepting proxy without
372 getting intercepted requests, error pages created from default templates
373 currently can result in client requests to config.privoxy.org on the
374 Internet which may not be desirable.
376 150) Add blocklistd support (https://man.netbsd.org/blocklistd.8).
378 151) Let the dok-tidy target work cross-platform without introducing
379 a ton of white-space changes that hide the content changes.
381 152) Fix CSS references in the website documentation.
382 For many pages p_doc.css is specified twice using different paths.
383 Usually at least one works, but not all of them do and the
384 duplicated requests are pointless even if they don't end up with
387 153) Catch SIGINT and use it to close the listen socket, serve
388 remaining connections and shut down. This would allow higher
389 uptime and make testing more convenient.
391 154) Underline links in docs and cgi pages. More precisely,
392 don't mess with the browser defaults for link underlining.
394 155) The sig_handler() shouldn't call log_error().
395 While it isn't known to cause actual problems in normal operation,
396 it's technically incorrect and causes crashes when running in
399 156) Reject socks requests with an explicit error message similar
400 to the one used for ftp. Motivation:
401 https://lists.privoxy.org/pipermail/privoxy-users/2017-March/000195.html
403 158) Use a single thread to wait for new requests on reused client connections.
404 Currently the thread that handles the first request on a connection
405 stays responsible for the client connect until it gets closed.
406 In case of lots of idle connections lots of waiting threads are used.
407 While it's conceivable that this ineffiency is irrelevant from a
408 performance point of view, using a single thread should reduce Privoxy's
409 memory footprint a bit which may be noticeable in case of multi-user setups
410 with hundreds of idle connections.
412 161) Properly support requests with chunked transfer-encoding with https inspection.
414 162) When https inspecting, delete generated keys and certificates if
415 the connection to the destination could not be established.
416 Makes silly DoS attacks slightly more complicated.
418 163) Use subdirectories in the certificate-directory to lower the number
419 of files per directory.
421 165) Add a max-connections-per-client directive.
423 167) Set up a public Privoxy-Filter-Test instance.
425 168) Add a privacy policy.
427 169) Preserve all relevant copyright and license statements in binary
428 packages we distribute.
430 170) Serve the ca-cert-file through the CGI interface so clients
431 can conveniently import it (insecurely).
433 171) Create a "view page using Privoxy" website where users can input
434 an URL and get a screenshot of a browser fetching the URL
437 172) Create a public git repository for Privoxy-Filter-Test.
439 173) Document Privoxy's governance model.
441 174) Let the Tor Onion Service for the privoxy website
442 serve gitweb and the git repository as well.
444 175) Add more screenshots to the documentation and website.
446 176) Find a new fiduciary sponsor as a replacement for Zwiebelfreunde e.V.,
447 so that we can continue to receive tax-deductible donations in Europe.
449 177) Support https-inspection for intercepted requests.
451 178) Warn on http://config.privoxy.org/client-tags if a Tag name
452 hasn't at least one matching action section.
454 179) Add a add-server-header{} action to add headers to the response
455 sent to the client (including responses generated by Privoxy itself).
457 180) Add support for GnuTLS.
459 181) Allow to upgrade an http request to https behind the
460 client's back using a client-header filter.
462 182) Before enforcing the client-header-order, check that the
463 client headers actually need sorting. Should reduce log
464 messages and memory allocations.
466 183) Properly deal with proxy responses that arrive in multiple pieces
467 when https inspecting while using a forwarding proxy.
469 184) Add support for wolfSSL. Work in progress, expected to be
470 committed after the 3.0.34 release. Funded with donations
471 made to the Privoxy project.
473 185) The mbedTLS and OpenSSL versions of generate_host_certificate()
474 should only be called when necessary and the check should be
475 done without holding the certificate mutex.
477 186) Privoxy should handle "OPTIONS *" requests properly.
479 187) There should be a convenient way to see the versions of
480 the libraries Privoxy is using.
482 188) In the windows config.txt file, add the line
483 user-manual ./doc/user-manual/
488 189) Bring back binary packages for macOS, preferably for both Intel and M1.
489 The first step would be getting at least one build system, either
490 donated or bought with donations earmarked for this.
491 Interested donors: 0.
493 190) The socks5 authentication code should send user name an password
494 seperately or we should increase the cbuf size to allow longer
495 user names and passwords.
497 191) The cipher-list directive should be split into cipher-list-server
498 and cipher-list-client.
500 192) The client TLS contexts should probably be shared among threads
501 to spend less time and memory loading the root certificates.
503 193) Use SHA256 instead of MD5 for the host hash used when generating file
504 names for host certificates and keys.
506 194) There should be a way to force gif deanimation if the server does not
507 declare the content as gif.
509 195) We should probably cache the server TLS contexts.
511 196) Investigate if it's worth adding an optional mutex for the CGI handler.
512 Could reduce memory use and increase performance on single core systems
515 197) Investigate if parts of Privoxy should get optional replacements
518 198) Add a config directive that prevent's IP addresses from being logged
519 (when logging is enabled).
521 199) In actions.c the "#define DEFINE_ACTION_ALIAS 0" lines should probably
522 be changed to "#undef DEFINE_ACTION_ALIAS" or removed.
524 200) Add a config directive that causes Privoxy to remove all
525 host certificates before exiting.
527 201) Add an action to change the trusted-cas-file for a section.
528 This should be useful in countries where a person-in-the-middle
529 attack is known to happen on some domains but should not be tolerated
530 on others. It would also allow to limit the accepted CA certificates
531 for given domains instead of accepting all that are specified with
532 the trusted-cas-file directive.
534 202) Allow to use multiple log files with different debug settings.
535 Suggested by Jonathan Marten in FR#607.
537 203) Add HTTP/2 support.
539 204) Make the Privoxy website available over IPv6.
541 205) Document how commit messages should look like.
543 ##########################################################################
545 Hosting wish list (relevant for #53)
550 - Mailinglists (Mailman with public archives preferred)
551 - Webspace (on a Unix-like OS that works with the webserver targets
554 - Commit mails (preferably with unified diffs)
556 (Unsorted) details to look at when evaluating hosters:
558 1. Preferably no third-party ads and trackers.
559 External images, CSS and JavaScript may count as trackers
560 but texts like "supported by company XYZ" may be acceptable.
562 2. JavaScript should be optional or not used at all.
564 3. Services we don't need shouldn't be enabled anyway.
565 (We currently don't use Web forums, wikis, surveys etc.)
567 4. It would be preferable if the hoster didn't have a bad track
568 record as far as user experience, security and privacy are
569 concerned and if the terms of service are "reasonable" and
570 haven't changed too often in the past. Updates in the past
571 should have been improvements and not regressions.
573 5. It would be preferable if most of the server administration
574 is done by a trusted third-party (or at least not a lot of work
577 6. The server(s) should be located in a country with laws we can
578 understand and follow (or at least not unintentionally violate).
580 7. A server location in a country with some kind of due process
581 and strong data protection laws (at least on paper) would be
584 8. Given that Privoxy is a free software project it would be
585 preferable if the hoster would use free software where possible.
587 9. Migrating away from the hoster in the future without losing
588 any important data should be possible without writing web