• last updated 20 hours ago
Constraints
Constraints: committers
 
Constraints: files
Constraints: dates

New API function: ad_docbook_xml_to_html

This function converts DocBook XML as used in the OpenACS

documentation to HTML. This is not a full implementation of all

possible DocBook markup, but just a subset sufficient for rendering a

substantial subset of the OpenACS documentation.

The function can be used via

set html [ad_html_text_convert -from application/docbook+xml -to text/html $text]

or

set txt [ad_html_text_convert -from application/docbook+xml -to text/plain $text]

fixedm typo in comment

rework register_icanuse "domDoc asTclValue" since w cannot use apm_version_names_compare during startup

Extend test suite

make use of new NaviServer command: ns_joinurl

the implementation provides a fallback when used with older versions of

NaviServer

fix comment of per_request_cache

generalized handling of nx classes and objects not defined by OpenACS

  1. … 1 more file in changeset.
added inline documentation for API browser

improved spelling

added inline documentation for objects for the API browser

added inline documentation for API browser

added inline documentation for API browser

exclude more content provided from outside of OpenACS

exclude low-level slot classes

improve spelling

provide full link to entities with missing text documentation

extended documentation for API browser

remove objects and classes defined in NaviServer from the listing

reduced reported error cases by ~1500 by omitting slot objects

Use tdom 0.9.5 feature when available

Prefer the builtin json2dict implementation of tDOM when available

improved inline documentation

improved API documentation

fixed test cases and improve documentation, added :util::block_request

Added code to skip suspicious looking query variables

On openacs.org, we are experiencing numerous requests with

multiply very long and strange query variables like in the example

below. So far, it is not clear, whether these requests are the

consequence of a double encoding or a deliberate attack. Many (most)

of the requests contain the query variable names containing the

(decoded) pattern "*amp;*".

This is a relatively new phenomenon. I cannot exclude that this is a

bug introduced lately in OpenACS, or a bug in an external bot, or

whatever. The problem with these query variables is that OpenACS

propagates these further, e.g., when updating query variables in

ad_dimensional, via export_vars, or return_urls.

Since OpenACS never uses these query-variables, these can be safely

skipped, without loosing functionality in OpenACS. It is possible to

construct examples, where skipping such variables can change the

semantics. Therefore, the change introduces a single function

util::suspicious_query_variable where in case of problems, the

skipping feature can be deactivated.

GET /api-doc/proc-browse?amp%3bamp%3bamp%3bamp%3bamp%3bamp%3bamp%3bamp%3bamp%3bamp%3bamp%3bamp%3bamp%3bamp%3btype=All&amp%3bamp%3bamp%3bamp%3bamp%3bamp%3bamp%3bamp%3bamp%3bamp%3bamp%3bamp%3bamp%3bamp%3bamp%3bamp%3bamp%3bamp%3bamp%3bamp%3bamp%3bamp%3bamp%3bamp%3bamp%3bamp%3bamp%3bamp%3bamp%3bamp%3bamp%3bamp%3bamp%3bamp%3bamp%3bamp%3bamp%3bamp%3bamp%3bamp%3bamp%3bamp%3bamp%3bamp%3bamp%3bamp%3bamp%3bamp%3bamp%3bamp%3bamp%3bamp%3bamp%3bamp%3bamp%3bamp%3bamp%3bamp%3bamp%3bamp%3bamp%3bamp%3bamp%3bamp%3bamp%3bamp%3bamp%3bamp%3bamp%3bamp%3bamp%3bamp%3bamp%3bamp%3bamp%3bamp%3bamp%3bamp%3bamp%3bamp%3bamp%3bamp%3bamp%3bamp%3bamp%3bamp%3bamp%3bamp%3bamp%3bamp%3bamp%3bamp%3bamp%3bamp%3bamp%3bamp%3bamp%3bamp%3bamp%3bamp%3bamp%3bamp%3bamp%3bamp%3bamp%3bamp%3bamp%3bamp%3bamp%3bamp%3bamp%3bamp%3bamp%3bamp%3bamp%3bamp%3bamp%3bamp%3bamp%3bamp%3bamp%3bamp%3bamp%3bamp%3bamp%3bamp%3bamp%3bamp%3bamp%3bamp%3bamp%3bamp%3bamp%3bamp%3bamp%3bamp%3bamp%3bamp%3bamp%3bamp%3bamp%3bamp%3bamp%3bamp%3bamp%3bamp%3bamp%3bamp%3bamp%3bamp%3bamp%3bamp%3bamp%3bamp%3bamp%3bamp%3bamp%3bamp%3bamp%3bamp%3bamp%3bamp%3bamp%3bamp%3bamp%3bamp%3bamp%3bamp%3bamp%3bamp%3bamp%3bamp%3bamp%3bamp%3bamp%3bamp%3bamp%3bamp%3bamp%3bamp%3bamp%3bamp%3bamp%3bamp%3bamp%3bamp%3bamp%3bamp%3bamp%3bamp%3bamp%3bamp%3bamp%3bamp%3bamp%3bamp%3bamp%3bamp%3bamp%3bamp%3bamp%3bamp%3bamp%3bamp%3bamp%3bamp%3bamp%3bamp%3bamp%3bamp%3bamp%3bamp%3bamp%3bamp%3bamp%3bamp%3bamp%3bamp%3bamp%3bamp%3bamp%3bamp%3bamp%3bamp%3bamp%3bamp%3bamp%3bamp%3bamp%3bamp%3bamp%3bamp%3bamp%3borderby=name&type=All&amp%3btype=All&amp%3bamp%3btype=All&amp%3bamp%3bamp%3btype=All&amp%3bamp%3bamp%3bamp%3btype=All&amp%3bamp%3bamp%3bamp%3bamp%3btype=All&amp%3bamp%3bamp%3bamp%3bamp%3bamp%3btype=All&amp%3bamp%3bamp%3bamp%3bamp%3bamp%3bamp%3btype=All&amp%3bamp%3bamp%3bamp%3bamp%3bamp%3bamp%3bamp%3btype=All&amp%3bamp%3bamp%3bamp%3bamp%3bamp%3bamp%3bamp%3bamp%3bamp%3bamp%3bamp%3bamp%3bamp%3bamp%3bamp%3bamp%3bamp%3bamp%3bamp%3bamp%3bamp%3bamp%3bamp%3bamp%3bamp%3bamp%3bamp%3bamp%3bamp%3bamp%3bamp%3bamp%3bamp%3bamp%3bamp%3bamp%3bamp%3bamp%3bamp%3bamp%3bamp%3bamp%3bamp%3bamp%3bamp%3bamp%3bamp%3bamp%3bamp%3bamp%3bamp%3bamp%3bamp%3bamp%3bamp%3bamp%3bamp%3bamp%3bamp%3bamp%3bamp%3bamp%3bamp%3bamp%3bamp%3bamp%3bamp%3bamp%3bamp%3bamp%3bamp%3bamp%3bamp%3bamp%3bamp%3bamp%3bamp%3bamp%3bamp%3bamp%3bamp%3bamp%3bamp%3bamp%3bamp%3bamp%3bamp%3bamp%3bamp%3bamp%3bamp%3bamp%3bamp%3bamp%3bamp%3bamp%3bamp%3bamp%3bamp%3bamp%3bamp%3bamp%3bamp%3bamp%3bamp%3bamp%3bamp%3bamp%3bamp%3bamp%3bamp%3bamp%3bamp%3bamp%3bamp%3bamp%3bamp%3bamp%3bamp%3bamp%3bamp%3bamp%3bamp%3bamp%3bamp%3bamp%3bamp%3bamp%3bamp%3bamp%3bamp%3bamp%3bamp%3bamp%3bamp%3bamp%3bamp%3bamp%3bamp%3bamp%3bamp%3bamp%3bamp%3bamp%3bamp%3bamp%3bamp%3bamp%3bamp%3bamp%3bamp%3bamp%3bamp%3bamp%3bamp%3bamp%3bamp%3bamp%3bamp%3bamp%3bamp%3bamp%3bamp%3bamp%3bamp%3bamp%3bamp%3bamp%3bamp%3bamp%3bamp%3bamp%3bamp%3bamp%3bamp%3bamp%3bamp%3bamp%3bamp%3bamp%3bamp%3bamp%3bamp%3bamp%3bamp%3bamp%3bamp%3bamp%3bamp%3bamp%3bamp%3bamp%3bamp%3bamp%3bamp%3bamp%3bamp%3bamp%3bamp%3bamp%3bamp%3bamp%3bamp%3bamp%3bamp%3bamp%3bamp%3bamp%3bamp%3bamp%3bamp%3bamp%3bamp%3bamp%3bamp%3bamp%3bamp%3bamp%3bamp%3bamp%3bamp%3bamp%3bamp%3bamp%3bamp%3bamp%3borderby=name&amp%3bamp%3bamp%3bamp%3bamp%3bamp%3bamp%3bamp%3bamp%3btype=Private&amp%3bamp%3bamp%3bamp%3bamp%3bamp%3bamp%3bamp%3bamp%3bamp%3btype=All&amp%3bamp%3bamp%3bamp%3bamp%3bamp%3bamp%3bamp%3bamp%3bamp%3bamp%3btype=All&amp%3bamp%3bamp%3bamp%3bamp%3bamp%3bamp%3bamp%3bamp%3bamp%3bamp%3bamp%3btype=All&amp%3bamp%3bamp%3bamp%3bamp%3bamp%3bamp%3bamp%3bamp%3bamp%3bamp%3bamp%3bamp%3bamp%3bamp%3bamp%3bamp%3bamp%3bamp%3bamp%3bamp%3bamp%3bamp%3bamp%3bamp%3bamp%3bamp%3bamp%3bamp%3bamp%3bamp%3bamp%3bamp%3bamp%3bamp%3bamp%3bamp%3bamp%3bamp%3borderby=name&amp%3bamp%3bamp%3bamp%3bamp%3bamp%3bamp%3bamp%3bamp%3bamp%3bamp%3bamp%3bamp%3btype=All HTTP/1.1" 200 62378 "" "Mozilla/5.0 (Windows NT 10.0; Win64; x64) AppleWebKit/605.1.15 (KHTML, like Gecko; compatible; FriendlyCrawler/1.0) Chrome/120.0.6099.216 Safari/605.1.15" "1729029614.331581 0.109805 0.000434 0.004026 0.215927

  1. … 6 more files in changeset.
make skipping reason explicit

Finish array reform in this proc

Rework array idioms

Formatting changes

Prefer unset to array unset when no pattern matching is needed

See https://wiki.tcl-lang.org/page/Dict+VS+Array+Speed

No need to fiddle in the ad_conn array if we just reset it