PNG
IHDR ; IDATxܻn0K
)(pA 7LeG{ §㻢| ذaÆ
6lذaÆ
6lذaÆ
6lom$^yذag5 bÆ
6lذaÆ
6lذa{
6lذaÆ
` }HFkm,mӪôô! x|'ܢ˟;E:9&ᶒ}{v]n&6
h_tڠ͵-ҫZ;Z$.Pkž)!o>}leQfJTu іچ\X=8Rن4`Vwl>nG^is"ms$ui?wbs[m6K4O .4%/ bC%t Mז -lG6mrz2s%9s@-k9=)kB5\+͂Zsٲ Rn~GRC wIcIn7jJhۛNCS|j08yiHKֶۛkɈ+;Sz L /F *\Ԕ#"5m2[S =gnaPeғL
lذaÆ
6l^ḵaÆ
6lذaÆ
6lذa;
_ذaÆ
6lذaÆ
6lذaÆ
R IENDB`
aNc @ s d Z d d l Z d d l Z d d l Z d d l Z d d l Z d d l Z d d l Z d d l m Z d d l
m Z d d g Z d e
f d YZ d Z d Z d
e
f d YZ d d
d Z d S( se
Middleware that profiles the request and displays profiling
information at the bottom of each page.
iN( t StringIO( t responset ProfileMiddlewaret profile_decoratorc B s/ e Z d Z d Z d d d d Z d Z RS( s
Middleware that profiles all requests.
All HTML pages will have profiling information appended to them.
The data is isolated to that single request, and does not include
data from previous requests.
This uses the ``hotshot`` module, which affects performance of the
application. It also runs in a single-threaded mode, so it is
only usable in development environments.
sW clear: both; background-color: #ff9; color: #000; border: 2px solid #000; padding: 5px;s profile.log.tmpi( c C s. | | _ t j | _ | | _ | | _ d S( N( t appt threadingt Lockt lockt log_filenamet limit( t selfR t global_confR R ( ( s7 /usr/lib/python2.7/site-packages/paste/debug/profile.pyt __init__& s c
s g g d f d f d } j j z+t j j } | j d j d d z | j | Wd | j Xd j
d } t j | d } | d k s | j
d r g St j j j } | j | j d d
t | j j } t | j j } d j t j | t j | f 7 g SWd j j Xd S( Nc s* j | | g | | | j S( N( t extendt append( t statust headerst exc_info( t bodyt catch_responset start_response( s7 /usr/lib/python2.7/site-packages/paste/debug/profile.pyt replace_start_response1 s c sG j } z j | Wd t | d rB | j n Xd S( Nt close( R R
t hasattrR ( t app_iter( R t environR R
( s7 /usr/lib/python2.7/site-packages/paste/debug/profile.pyt run_app5 s
t URLt PATH_INFOt i s content-types text/htmlt timet callss %s
%s
( t NoneR t acquiret hotshott ProfileR t addinfot gett runcallR t joinR t header_valuet
startswitht statst loadt
strip_dirst
sort_statst capture_outputt print_statsR t
print_callerst stylet cgit escapet release(
R
R R R t profR t content_typeR* t outputt output_callers( ( R R R R R
R s7 /usr/lib/python2.7/site-packages/paste/debug/profile.pyt __call__. s4
& N( t __name__t
__module__t __doc__R1 R R R9 ( ( ( s7 /usr/lib/python2.7/site-packages/paste/debug/profile.pyR s c O sC t } t j } | t _ z | | | Wd | t _ X| j S( N( R t syst stdoutt getvalue( t funct argst kwt outt
old_stdout( ( s7 /usr/lib/python2.7/site-packages/paste/debug/profile.pyR. V s
c s/ j d r d } | S f d } | S( s
Profile a single function call.
Used around a function, like::
@profile_decorator(options...)
def ...
All calls to the function will be profiled. The options are
all keywords, and are:
log_file:
The filename to log to (or ``'stdout'`` or ``'stderr'``).
Default: stderr.
display_limit:
Only show the top N items, default: 20.
sort_stats:
A list of string-attributes to sort on. Default
``('time', 'calls')``.
strip_dirs:
Strip directories/module names from files? Default True.
add_info:
If given, this info will be added to the report (for your
own tracking). Default: none.
log_filename:
The temporary filename to log profiling data to. Default;
``./profile_data.log.tmp``
no_profile:
If true, then don't actually profile anything. Useful for
conditional profiling.
t
no_profilec S s | S( N( ( R@ ( ( s7 /usr/lib/python2.7/site-packages/paste/debug/profile.pyt decorator s c s f d } | S( Nc s t | | S( N( t DecoratedProfile( RA RB ( R@ t options( s7 /usr/lib/python2.7/site-packages/paste/debug/profile.pyt replacement s ( ( R@ RI ( RH ( R@ s7 /usr/lib/python2.7/site-packages/paste/debug/profile.pyRF s ( R% ( RH RF ( ( RH s7 /usr/lib/python2.7/site-packages/paste/debug/profile.pyR b s
" RG c B s8 e Z e j Z d Z d Z d Z d Z RS( c K s | | _ | | _ d S( N( R@ RH ( R
R@ RH ( ( s7 /usr/lib/python2.7/site-packages/paste/debug/profile.pyR s c O s< | j j z | j | j | | SWd | j j Xd S( N( R R! t profileR@ R4 ( R
RA RB ( ( s7 /usr/lib/python2.7/site-packages/paste/debug/profile.pyR9 s
c O se | j } | j d d } t j | } | j d | j | | | | j d ro | j d | d n d } zK t j } y | j | | | } Wn t j
} n Xt j }
Wd | j Xt j j
| } t j | | j d t r | j n | j | j d d | j d
d } t | j | }
t | j | } | j d } | d k rt j } nV | d k rt j } n> t | d } | j d d d | j d t j d | j d | j | | | | j d |
| | j |
| j | | d k r@| j n | ra| d | d | d n | S( NR s profile_data.log.tmps
Function Callt add_infos
Extra infoR, R- R R t
display_limiti t log_filet stderrt -R> t as
%s
i< s Date: %s
s %cs Function call: %s
s Wall time: %0.2f seconds
i i i ( s times calls( Ns stderr( RO s stdout( NRO s stdouts stderr( RH R% R" R# R$ t format_functionR R R& R= R R R* R+ t ost unlinkt TrueR, R- R. R/ R0 RN R> t opent writet strftime( R
R@ RA RB t opst
prof_filenameR5 R t
start_timet resultt end_timeR* RL R7 R8 t output_filet f( ( s7 /usr/lib/python2.7/site-packages/paste/debug/profile.pyRJ sX
c O s_ t t | } | j g | j D] \ } } d | | f ^ q" d | j d j | f S( Ns %s=%rs %s(%s)s , ( t mapt reprR
t itemsR: R' ( R
R@ RA RB t kt v( ( s7 /usr/lib/python2.7/site-packages/paste/debug/profile.pyRQ s 0( R: R; R R R R R9 RJ RQ ( ( ( s7 /usr/lib/python2.7/site-packages/paste/debug/profile.pyRG s
1s profile.log.tmpi( c C s" t | } t | d | d | S( s
Wrap the application in a component that will profile each
request. The profiling data is then appended to the output
of each page.
Note that this serializes all requests (i.e., removing
concurrency). Therefore never use this in production.
R R ( t intR ( R R R R ( ( s7 /usr/lib/python2.7/site-packages/paste/debug/profile.pyt make_profile_middleware s ( R< R= RR R" t
hotshot.statsR R2 R t cStringIOR t pasteR t __all__t objectR R. R RG Re ( ( ( s7 /usr/lib/python2.7/site-packages/paste/debug/profile.pyt s A ,I