CVS log for parser3/src/main/pa_globals.C

Up to [parser3project] / parser3 / src / main

Request diff between arbitrary revisions - Display revisions graphically

Keyword substitution: kv
Default branch: MAIN

Copyright year updated, websites links changed to https://

pa_GC_abort_func added and used (implements feature #1283)

fixes for --with-build-warnings

classes.awk is useless, preparing to delete it

Copyright year updated

Adopted patch from Yavor Doganov <yavor@gnu.org> to add pcre2 library support (implements feature #1236)

copyright, authors actualized

^memory:auto-compact(N) added (implements feature #1209)

Copyright year updated

GC_dont_gc replaced

GC_java_finalization is depricated

thread request check added

no longer required due to -MD -> -MT change (releated to feature #1200)

As we log allocation errors, we don't want default gc warnings (without timestamp and URI).

GC_dont_gc=1 moved to pa_globals_init

setup_hex_value() removed, static array initialization used (optimization)

pa_socks_init/pa_socks_done moved inside pa_globals_init (optimization)

simplified to XML_STATIC

libpcre now linked same way as other libs

x64 libs support

PA_DEBUG_DISABLE_GC works again, pa_gc_* method removes, libcord also can work without libgc

Copyright year: 2015->2017

vs2015 compilation fixes

SYMBOLS_CACHING define implemented (related to feature #1026)

minor cleanup

Copyright year updated

warning war (-pedantic flag)

spelling fixed

cleanup

PA_DEBUG_DISABLE_GC compilation fix

PA_RELEASE_ASSERTS removed

pa_dlinit added for lt_dlinit to be called once and lt_dlexit called at right place (related to issue #925)

ident now works under Linux + ident displays filenames (closes issue #818)
Copyright updated

the rest converted to THREAD_LOCAL usage

THREAD_LOCAL defined and used

- die instead of abort

- moved to new PCRE library

- pcre now everywhere in separate namespace

. killing gdome [what a relief]

. started killing gdome

. merged latest fixes from 3.2.0

. less warnings

//20051130 trying to remove this, author claims that fixed a lot there // 20040920 for now both workarounds needed. wait for new libxml/xsl versions

2005

. 2005 ;)

debugger help: PA_RELEASE_ASSERTS enables release asserts

libxml has bugs: it calls xmlMallocAtomic somewhere where it should have called xmlMalloc
and it calls xmlFree when it should have not called it.
inserted two workarounds: xmlMallocAtomic implemented as xmlMalloc, and xmlFree just ignored.
put away a testcase, maybe someday libxml author would fix all that.
until that day: we have SLOW garbage collecting when many xml objects are alive
hint: do ^memory:compact[] before xdoc::create

bugfix: sql connections now are disconnected

bugfix: xml errors copy from library now cleared [were not cleared properly]

change: .xsl.stamp now not used, introducing dependencies timestamp check

2004

beauty: typo caused gc.log warnings with debug version of gc.dll

migrated to latest libxml[2.6.5] and libgdome[1.1.2]
simplified lib build processes

continued changes to reflect libxml xsl new folders

started changes to reflect libxml xsl new folders

beauty: bigger buffer for xml-related errors

bugfix: removed false warning [warning: unreported xmlGenericErrors]

bugfix(solaris): removed dependency on object initialization order [one more place]

bugfix: win32: compile prob

simplification: xml errors to hash<thread_id,error>
reorganization: xml io moved  to pa_xml_io.C/h

static const char * const IDENT

turned on warnings level4 on all projects (except libltdl=off & libsdbm=level3)
found several unitialized vars

when xml memory allocator returns 0, just die.
http://i2/tasks/edit/?id=4570798492410259445

cvs: getting rid of win32xml pseudo project

moved tempate_gc to HEAD

relative paths to xml&gc libs

few paths for libxml debug/release changes

minor style changes

introducing
$f[^file::load[binary;http://...]]
$f.tables

xml memory debugging functions (ifdefed)

started digging on double free.
on win32 found that that's perfectly normal.

pcre memory management changed to use GC memory

2.5.6 version of libxml allows to install xmlMallocAtomic [author agreed to my suggestion], used that

append_array_limit_sense_change
locate accepts options same as create
new option: $.reverse(1)   [do not work in table::create]

no message

exif mem leak
http://i2/tasks/edit/?id=4480590323629807263

links to xml libs made relative, no need to unpack parser3project to /parser3project.

*** empty log message ***

globals.c: gc_substitute_memory_management_functions
+installed CORD_oom function

moved memory debugging to global level: to pa_gc_malloc

xml memory debugging showed no errors. todo: debug parser memory

prepared xml-static configuration

started xml lib memory debugging, simple checks failed:
it seems that library does realloc(bad ptr), and recording those ptrs in heap [for debugging] was bad:
heap situation changes = everything works fine

no message

XML memory funcs to GC works in debug, but fails in release
todo: fix that

forgot to merge XML memory funcs replacement from gc branch

untaint.C L_SQL

started XML

2*2 worked :)

pa_sql_driver_manager.C compiled

started porting to gc:
PA_Object done
lots of replacements also

libgc attempt

xml: test: 107 bug fixes

xnode.C compiled

xml: exceptions

SAPI::abort << abort.  die now just exits

op compiled

date compiled

removed stupid value_includes.h

merged changes from 1.149-1.150 (stupid name conflicts)

vmail compiled

most pa_request compiled

grammar compiled

vmethodframe, pa_request.h compiled

vfile compiled

vcookie compiled

vrequest,vdouble,vint compiled

globals compiled

pa_charsets compiled

more patched

resurrected pool in new sense: now it's factory, producing&accounting memory chunks for read[autofree] buffers

aint that easy

2002->2003

http://
introducing
	$.any-status(1)

localized pa_xmlFileRead/Close

moved to latest xml lib versions, changed patches.
      libxml2 >= 2.5.1 [ftp://xmlsoft.org/libxml2-2.5.1.tar.gz]
      libxslt >= 1.0.23 [ftp://xmlsoft.org/libxslt-1.0.23.tar.gz]
      gdome2 >= 0.7.2 [http://gdome2.cs.unibo.it/tarball/gdome2-0.7.2.tar.gz]

optimized-xml

optimized-as-is

$response:download

response:body[file]  if file nows it's name
now returns content-disposition: >>attachment<<

no message

no message

http://
and image const void related probs fixed

table/xdoc/file::load now understand http:// prefix and additional params, sample:
$rates[^xdoc::load[http://www.cbr.ru/scripts/XML_daily.asp?date_req=02/03/2002;
   $.USER-AGENT[parser3]
]]

$image.exif support
$image.exif.DateTime & co

table.sql hash::sql flag to allow duplicate keys [$.distinct(1/0)]
first record taken [were last]

prepared:
	// validate each document after load/create (?)
	//xmlDoValidityCheckingDefaultValue = 1;

year column in month calendar [week year]

$file.content-type [instead of .mime-type]
now works for file::stat also

month calendar week column [ISO 8601 Week Numbers]

$Date: now$

ident.C* removed

win32xml win32mailreceive pseudomodules paths now

uue to separate file
introducting $message:received
^mail:send rewritten with backward comp

parser3.conf now one and only

@rootconf [were @conf]

@conf

fixed .dsp-s along with reorganized cvs modules dirs structure

reorganized cvs modules

Table name2number field now &.
main_method_name now on Request pool rather on global_pool [Junction+VJunction created on same pool as name = were created on global pool, causing mem leaks]

Table name2number field now &.
main_method_name now on Request pool rather on global_pool [Junction+VJunction created on same pool as name = were created on global pool, causing mem leaks]

removed 100*40 bytes per request memory leak [nonpool malloc in globals]

removed 100*40 bytes per request memory leak [nonpool malloc in globals]

fixed match table template

String::match table columns globalized, not created @ each ^match anymore
economy: sizeof(Array)+space on 3+x cells=24+ bytes per ^match

removed Value::fname

switch in hash constructor fixed [were problems with using of stacked wwrapper after it's death]

^cache[file][date]{
   ^cache(seconds)
   ^cache[date]
}

introducing ^try

isapi site config beside .dll

removed hashfile support for now

extern "C" was removed too fast

removed unneded
#ifdef __cplusplus
}
#endif

win32xml now contains gnome xml libs
parser .dsp-s changed accordingly

name spelling

2002

few forgotten ifdef XML added

http://localhost/abc -> $ENV{DOCUMENT_ROOT}/abc | ./abc

xdoc::load/set entities substituted.
	// 2. when dom tree with entites goes under transform text nodes
	// got [erroreosly] cut on first entity occurance

%.20E now default float format
this is for regretful twice-converting here:
$a(double value) when they write double falue  they convert it to string first,
thus %g were losing precesion.

todo: think up some way to remove double->string string->double twice-converting

xmlSubstituteEntitiesDefault(1) now global, reasons in comment inside

xslt linked dynamically. [mistekenly were statically linked]

removed debug xslt messages

xsl stylesheet load: parsed entities, this helps compiling stylesheet properly.
example:
this refused to work,
            <b>&laquo;<xsl:value-of select="@title" />&raquo;</b>
stranly worked only this way:
            <b>&#171;<xsl:value-of select="@title" />&raquo;</b>

docs says "set it to 1", never going into details, so I decided to try NOT to do that

autoconf gnome

xslt generic error 1

xml generic errors 0

xslt transform0.
TODO:
  use output options
  handle errors

xdoc
^set
^string with glib works 0

renamed gdome to libgdome.dll

gnome libs just compiled in.
no refcounting
no xpath
no xslt
  yet

going away from xalan&xerces, started the process.
globals [initialization
pool [charset update
charset [transcodings
vxnode vxdoc [DOM calls

introducing Charset

removed unused defalts_name global

~Cache_managers

UL_CLEAN pieces now inherit UL_OPTIMIZED_BIT at wcontext.write

introducing estimating String::cstr_bufsize, 0

cache_managers, maybe-expiring every request
todo: apache 'status' class enabler

$status:db
fixed used not initialized in db_connection & db_table

started status parser class

debugging memory
hash key ref now

[Mon Oct 29 16:01:17 2001] total 17050496/202647
[Mon Oct 29 17:11:17 2001] total 16867948/202658

DB2 ifdefs

vc projects libdb ifdefed into globals.C

ifdefs so it compiled without db

hashfile 0

xalan&xerces multithread bug fixed. initialization&free on each thread!

untaint lang origins table fixed.
got rid of empty_string, which caused errors - it hasnt exception but somebody[value.bark] tried to throw it on it's pool

would make charset loading & caching mech now

now charsets table configuration variable is: CHARSETS

added ident.awk and main.dsp splitted to main+classes+types along to .am

merged with before_xml

zillions of #ifdef XML
created new projects for sql drivers [planning to remove sql drivers from main cvs project]
first, will join latest changes

Pool::transcode 0

started dnode+dom

dnode
    DOM methods:
    $node.name
    $node.value
    ...others...

    ^node.xpath[/rates/USD] = hash
        $hash[
        	$.0[node0]
        	$.1[node1]
        ]

dom.xslt  stylesheet compiled&cached

^dom:set{<some>xml</some>} default language XML
languages+=xml

now there's special hash key '_default' [instead of ^_default method]
$hash[
	$.a[1]
	$.b[2]
	$._default[xx]
]
$hash.c

!^int/double:sql{query}[[$.limit(2) $.offset(4) $.default(0)]]
string, table

class_path

memset dictionary minor bug fixed

found another multithread bug in op.C (last, last global var killed) :)

found another multithread bug in op.C (last, last global var killed) :)

speeded up typo-html replacements. introducing Dictionary with first-char caching

match replace code context moved to implicit $match context

$ORIGINS(1) output tracing mode

*/
static const char *RCSId="$Id: double.C,v 1.31 2001/06/28 07:41:59 parser Exp $";

static char *RCSId="$Id$";

z

moved some configured data to request::classes_conf
moved some string crations from globals to M...
[works]

moved some configured data to request::classes_conf
moved some string crations from globals to M...

Methoded reorganized. todo: methoded-configure

beautifying just compiled. todo: debug, configure

beautifying -100

started beautifying

$MAIN:LOCALE

^unknown:int[]=0 double[]=0

gd dead end. switching to python imaging lib

image:measure -100 just compiled

smtp some consts

exec4. todo env

exec -2

problems with ^exec:cgi post data. they are already read by vform

smtp just compiled

mail:send -10 just compiled

table:dir 1

renamed 'root' to 'op'

@office

$LOCALE:ctype[Russian_Russia.1251]

connect&sql -1000 just compiled

string:match 0

string:match [search] -1 just compiled

content-disposition

main:post-process

random:generate

file:load autodetection of mime-type by user-file-name

moved ::save to pa_table.
disabled @auto invocation in ^process

html-typo - moved all processing to table [except preliminary \r\n \r \n replacements to "\n"]

returned table originating. useful for reporting typo table problems origin. think would be useful somewhere else

typo & typo-default

html-typo 0
html-typo 1. noticed double default content-type prob. run cgi.cmd tomorrow first
html-typo sample. detected << problem: they become ltlt too fast

fixed default content-type allocation storage

z

thrown away freeze mech.
no danger now for you can't do ^request:process[@new-method[]
body]
trick anymore.
still remains $some_instance.process[@new-method[]
possibility]
but would leave it as a feature.
see index for sample

sapi made object

config of auto.p path: parser_root_auto_path parser_site_auto_path

shifted apache branch on main trunc

String::UL_

z

VRequest moved to :Value

VRequest moved to :VStateless_object

VResponse moved to :VStateless_object

VEnv is now :Value

VCookie is now :Value

\ to /  and ^file:save

file class just compiled

z

auto @brief

cookie class

$defautl[$content-type[$value[text/html] $charset[windows-1251]]

^taint 0

convinient string(pool, char *src, bool tainted) ctor

proper @exceptions params tainting

z

$response: 0  request::core exception rethrow

$request : query :uri

$result

vform fillfields just compiled

limits -1

request_info -1

vform_fields_fill

global handler, exceptions in parser3