I am trying to add everything I have done - but do not take this list for complete! - Olaf Noehring ==================================================================== + new / added - removed * changed (works different now) / checked / tested (tag ) search this file for the tags to get more information /* * following will be filled automatically by SubVersion! * Do not change by hand! * $LastChangedDate: 2005-05-17 17:18:57 +0200 (Tue, 17 May 2005) $ * @lastedited $LastChangedBy: olaf $ * $LastChangedRevision: 41 $ */ -------------- version 0.942.02 * indexoverview: corrected ugly output in "complete index overview" (Olaf) indexoverview complete: output now fine indexoverview short: overflow hidden if colums are too small, title tag added placed "protected status"-checkbox directly under file size (Olaf) * indexedit: placed "protected status"-checkbox directly under file size (Olaf) -------------- version 0.942.01 * prevent reading pages twice - once for meta, once for content (Manfred) * https fix (hopefully) (Toon) * protect from multiple runs (Toon) -------------- version 0.941 offical release -------------- version 0.941.13 (not public) * improved indexer call: now shows the domain in the first field of the site TSEP is installed, shows the absolute path in the second field (Toon) * changed indexer output: Only one line per page now. Complete index is still available on another page. (Toon) -------------- version 0.941.12 (not public) * minor change to logview: added a breaker div to make display in FF correct * updated install to take new fields into account (Toon) * updated language.php to have shorter entries in logview -------------- version 0.941.11 (not public) * logview: Title attribute to all entries. So if they are cut off, the user can still hover over them to read the complete entry * logview: If log is type 2 only the link itself is clickable. This allows to show the title better -------------- version 0.941.10 (not public) * updated documentation * fixed stopwords / regex: Spaces between words should now be shown correctly when a stopword was searched. This also fixed wrong number of results. (Manfred)) * updated logview: Stopwords and IPresolve can be filtered now filter stays when sorting is changed (putsortpicture.php updated for this as well) now, after removing criterias from content, stopwords, ipresolved all records are returned again (search file for "MySQL bug??" 3 occurences) added calendar removed | between multiple stopwords (change was in log.php) * updated CSS to make a nice logview.php * created new directory JS to hold javascript. This can not be under /include or /admin since the user would need a password if those directories are protected -------------- version 0.941.08 (not public) * added improved logging and logview: IP resolving (Toon, Eric) * updated documentation * updated ipfunctions.php, configfunctions.php to require files so that it should work even when open_basedir is in on * updated website * updated insert.sql -> DB Version number -------------- version 0.941.07 (not public) * fix UTF bug (Manfred made a workaround the MySQL UTF bug): ifIf Mysql >= 4.1 still the "normal" search is done. If Mysql < 4.1 AND the search-string DOES NOT contain 'special' chars (like e.g. "à") still the "normal" search is done. If Mysql < 4.1 AND the search-string DOES contain 'special' chars (like e.g. "à") tsep_search-table is read without using a where-clause. The Result is built by testing each record against a regex (built out of the search-string) respecting utf8. * changes language directories - to match better for multilanguage use: Now 4 charcters format: language 2 chars + "_" + variation 2 chars (en_US) -------------- version 0.941.06 (not public) * updated todo * added .htaccess to /include directory (deny from all) * added "ipresolved VARCHAR(250) NULL, stopwords TEXT NULL," to create sql from _log table -------------- version 0.941.05 (not public) * logview: basic filter function added (will be improved in future version): (Toon) Filter on logview type, searchterm, date/time, IP address Filter is active after choosing different sort order and through navigation -------------- version 0.941.04 (not public) * corrected configuration.php: path for languages can be 2-5 characters long now (e.g: pt_PT and pt_BR) * improved tsepinfo.php: Noew showing also MySQL version * added hint in install.php for dbpassword security * added CHARACTER SET utf8; to create.sql script to make sure UTF-8 is used in MySQL >=4.1 -------------- version 0.941.03 (not public) * fixed UTF-8 encoding bug. BIG thanks to Manfred here! * install.php: Passwortfield is now password (shows *) -------------- version 0.941.02 (not public) * removes language bug: $tsep_lng = array(); --> we need to pay attention that this does not happen again -------------- version 0.941.01 (not public) * removed $tsep_lng['language_iso_setting'] from languages file since all is done now with UTF-8 * changed sortortder of an entry in the indexer to make more sense * introduced groups in the indexer (needs new install) -------------- version 0.940 * introduced completely new way of translating TSEP * changed languase charset used by TSEP to UTF-8 * fixed log: search for "o'neil" is now correctly logged as "o'neil" and not anymore as "o\'neil" * added traces to search and indexer: ATTENTION: switch OFF the debugprint function in configuration.php !! * search.php : fixed error that a search word was too short is being output only if it was really to short (<4 characters) now corrected the search option: was +Apfel +Birne has searched: "Apfel" and "Birne" +Apfel+Birne has searched : "Apfel+Birne" now: +Apfel+Birne searches "Apfel" and "Birne" pages which can not be accesses by TSEP will be checked in a second way if they exist before outputting them in the search results output of search for "o'neil" is now correctly printed as "o'neil" and not anymore as "o\'neil" * indexer: fixed: out of memory bug when indexing and index was really large multiple spaces are removed from the index now much faster now (1200 files in 130 seconds) dir_exclude improved * updated all stopword list files to have comma seperated values * updated create.sql to create MyISAM tables (should be default but problems were reported) * changed database field in search table from text to longtext to be able to index (very)large html files * updated insert.sql: now the page rank can be html code again (hopefully) * htaccess protected directories should be indexed fine now * updated documentation: added: hint how to use stopword lists add a hint to words which are in more than 50% of the searches might not return anything! add hint: 3 character are mysql restriction make clearer what quick.htm and complete.htm are all about add hint: search from indexer is not logged hint: look at tsepsearch.php for details how to build your own search page -------------- version 0.939 * removed bug with TSEP path (now stored when opening configuration.php) * updated docs * updated todo -------------- version 0.938 + removed bug in IE on indexer.php -> mouse over links breaks layout, was CSS error + CSS corrected + TSEP path is saved automatically + added correct description in indexoverview to title and URL link + introduced tags: tsep:cmd:start, tsep:cmd:end, tsep:cmd:noindex + updated and split documentation * removed bug from phpcrawl4tsep + pages are now shown in the iso encoding used in the language.php (tag: ) * corrected english language.php - TRANSLATORS: please update the translations! + removed indexedit link from navigation area. + indexoverview: click on page title opens the indexedit, click on URL opens the page itself + created the putsortoder.php file in /include. Was in logview.php now we can use it in other files (indexoverview) Corrected the fuction to work this way (new $gotoURL parameter) + sort indexoverview + corrected polish language directory name (po now correct pl) * configuration: true / false should be dropdown boxes in configuration.php + groups in configuration + each field can have help * configuration of between how many results the user can choose: multiple values can be entered at once, no doubles can be entered * stopwords: can be entered comma seperated now; no doubles can be entered * starting parameters for indexer.php: indexer.php?startindexing - starts creating a new index for the current indexingprofile indexer.php?startindexing&profile=xxx - starts creating a new index for indexingprofile named xxx indexer.php?showcompleteindex - show complete index currently stored indexer.php - shows the maintain-screen + meta tags are saved now (was incorrect before). They are put into an extra field * new database structure - change tsep_config: formfieldvalue to different (larger fieldtype) + Indexingprofiles: + work with profiles from indexer.php + an indexentry can belong do multiple profiles + protect entry from being indexed automatically + indexing only for the chosen profile 1. delete all entries of the profile which are not protected 2. create new index of all matching (to that profile) files/directories with no overwriting of protected files If an entry is not protected it will be renewed even if it belongs to different profiles. * correct the search field in the indexer: When you search any special characters ( -> ä) from here it does not work + many other changes! -------------- version 0.935.5 (not public) + grab TSEP path for the installation and the indexer.php (no need to enter that manually anymore) * fixed: not really indexing files but telling us so (not really indexing again ;-) ) + local files are opened via local fopens. this should prevent firewall actions which disallows us to open files to read their contents * fixed call of css should now be correct -------------- version 0.935.3 (not public) * removed many superfluous tags in indexer.php this solved the problem of loosing configuration * removed "weird code" part from indexer + fixed error with outputting $tsep_lng['pages_found'] when that contained special characters ( "pages trouvées."), Thanks for the hint Alain + fixed SF bug 1082257: check excluded directories in the section of code that recursively calls search_get instead of the section , dir_exclude is being checked also in the read_dir-function * correct the number of pages shown after creating a new index * changed 0929 and 0925 was not public + fileextension-filter now can be maintained from within indexers "build new index"-page + fileextension-filter also made available for external datasupply via global var $TSEPextinclude * phpcrawl4tsep.php: AllowFollowMatch-definition now uses $TSEPextinclude -------------- version 0.926 (not public) * placed $start_time in indexer in the bedinning of the file * removed bug when accessing a directory * change "(c) TSEP" to "powered by TSEP" * updated documentation -------------- version 0.925.2.ON (not public) * updated french language from Alain (additional) * time limit possible (set to 1800) now if safemode is off (might be important for many files) tag: now: // if (!( $ct == 10 or $ct == 25 or ($ct % 50) == 0 )) // return; we are NOT spidering ourselves, but we let others spider for us! + add crawling / spidering along URL additional to opening files by fopen. The admin can choose which way. Also both can be done (indexing twice, but writing of course only one entry for each page to the database) : http://phpcrawl.sf.net and phpdig.net > >And one more thing. My translation is encoded in ISO-8859-2 but your scripts don't allow to change the encoding. You could add one more variable in you config file or, which I think would be better, a variable in the language file. This variable could be used to determine which encoding is to be used. > I do not really understand what you mean, please explain. Well. AFAIR the tag can't be changed in an easy way. That involves some problems because Polish special characters require ISO-8859-2 charset encoding. That's why I recommend adding a new variable to the config file or to every of the translated texts files. Example: in one of the files I've mentioned: $charset = 'ISO-8859-2'; and in every html header: echo('