iwla/ChangeLog

119 lines
4.2 KiB
Plaintext

v0.7 (17/03/2024)
** User **
Awstats data updated (7.9)
Improve page/hit detection
--display-only switch now takes an argument (month/year), analyze is not yet necessary
Add --disable-display option
Geo IP plugin updated (use of [ip-api.com](https://ip-api.com/))
Add _subdomains_ plugin
New way to display global statistics : with links in months names instead of "Details" button
Add excluded domain option
** Dev **
Remove detection from awstats dataset for browser
Don't analyze referer for non viewed hits/pages
Remove all trailing slashs of URL before starting analyze
Main key for visits is now "remote\_ip" and not "remote\_addr"
Add IP type plugin to support IPv4 and IPv6
Update robot detection
Display visitor IP is now a filter
Generate HTML part in dry run mode (but don't write it to disk)
Set lang value in generated HTML page
Add no\_referrer\_domains list to defaut_conf for website that defines this policy
Set count\_hit\_only\_visitors to False by default
** Bugs **
Flags management for feeds display
v0.6 (20/11/2022)
** User **
Replace track_users by filter_users plugins which can interpret conditional filters from configuration
Don't save all visitors requests into database (save space and computing). Can be changed in deufalt_conf.py with keep_requests value
Replace -c argument by config file. Now clean output is -C
Add favicon
Be more strict with robots : requires at least 1 hit per viewed page
Add a plugin to anonymize IP addresses (for public statistics)
Allow to merge feeds based on name regular expression with merge_feeds_parsers_list conf value
Add all_visits_enlight display plugin
Show last access information for feeds parsers
** Dev **
Update data from AWStats
Add Geo IP location information
Normalize URL before counting in stats
Update feed detector : check 'feed', 'rss' or 'atom' string in user agent
Move code to Python 3
Check for multimedia files using lower case
Don't launch robot analysis rules for feed parsers
** Bugs **
Fix a bug in feed filter (must not return, just break after manage of feed_name_analysed)
Display viewed and no viewed data for feeds (can be set as a robot too late)
v0.5 (15/04/2020)
** User **
Add --dry-run (-D) argument
Add more rules for robot detection :
More than ten 404 pages viewed
No page and no Hit
Pages without hit
New format for (not_)viewed pages/hits and bandwidth that are now recorded by day (in a dictionnary were only element 0 is initialized). Element 0 is the total. WARNING : not backward compatible with previous databases.
Sync data with awstat (develop branch : 7.7+)
Make backup before compressing (low memory servers)
Add top_pages_diff plugin
Add IP exclusion feature
** Dev **
Use cPickle instead of pickle
Don't save all robots requests (only first pass is kept) which allow to save a large amount of memory/disk space
Handle URLs with empty referer
Don't try to find search engine on robots
** Bugs **
Fix KeyError : geo attribute can not exists
Call post hook plugins even in display only mode
v0.4 (29/01/2017)
** User **
Remove crawlers from feed parsers
Add display only switch (-p)
Add robot bandwidth display plugin
** Dev **
** Bugs **
for robots, we have to use not_viewed_pages (feeds plugin)
gz files were not generated due to bad time comparison
Database compression could lead to altered files
v0.3 (12/04/2016)
** User **
Add referers_diff display plugin
Add year statistics in month details
Add analysis duration
Add browsers detection
Add operating systems detection
Add track users plugin
Add feeds plugin
Add _append feature to conf.py
Add hours_stats plugin
Add display/top_downloads_diff plugin
Can specify multiple files to analyze
Add reset feature
Add gz files support
Add -z option (don't compress databases)
Add own search enfines files
Do reverse DNS on feeds parsers
Add IPToGeo plugin
** Dev **
Add istats_diff interface
Sort documentation output
Add debug traces in robots plugin
Update awstats data
Remove double slashes at the end of URL
Remove final slashes for referrers
Add alt attribute for all img tag
** Bugs **
Forgot <body> tag
Bad UTC time computation
Hits/pages in the same second were not analyzed
Last day of month was skipped