Mercurial > hg > cc > cirrus_work

annotate lib/python/cc/ix.py @ 298:fdec28613df3 default

Find changesets by keywords (author, files, the commit message), revision number or hash, or revset expression.
robotstxt now working?
author Henry S. Thompson <ht@inf.ed.ac.uk>
date Mon, 05 May 2025 20:57:46 +0100
parents a4538bdfa92a
children 4759d6ac4625
Ignore whitespace changes - Everywhere: Within whitespace: At end of lines:
rev   line source
18
046dbe557911 write to tmp file implemented
Henry S. Thompson <ht@inf.ed.ac.uk>
parents:
diff changeset
1 #!/usr/bin/env python3
41
fa43c318749b correct comment
Henry S. Thompson <ht@inf.ed.ac.uk>
parents: 21
diff changeset
2 '''Extract response records from Common Crawl WARC-format files
18
046dbe557911 write to tmp file implemented
Henry S. Thompson <ht@inf.ed.ac.uk>
parents:
diff changeset
3 given length, offset and filename triples.
046dbe557911 write to tmp file implemented
Henry S. Thompson <ht@inf.ed.ac.uk>
parents:
diff changeset
4 Input one triple on command line, or
046dbe557911 write to tmp file implemented
Henry S. Thompson <ht@inf.ed.ac.uk>
parents:
diff changeset
5 triples from stdin as tab-delimited lines
046dbe557911 write to tmp file implemented
Henry S. Thompson <ht@inf.ed.ac.uk>
parents:
diff changeset
6 or complete cdx index lines.
046dbe557911 write to tmp file implemented
Henry S. Thompson <ht@inf.ed.ac.uk>
parents:
diff changeset
7 In all cases by 'filename' is meant crawlid/segmentid/type/filename
046dbe557911 write to tmp file implemented
Henry S. Thompson <ht@inf.ed.ac.uk>
parents:
diff changeset
8
046dbe557911 write to tmp file implemented
Henry S. Thompson <ht@inf.ed.ac.uk>
parents:
diff changeset
9 Note that if no output flag(s) is/are given, the whole WARC record will be output, more efficiently than would be the case if -whb is given.'''
046dbe557911 write to tmp file implemented
Henry S. Thompson <ht@inf.ed.ac.uk>
parents:
diff changeset
10
046dbe557911 write to tmp file implemented
Henry S. Thompson <ht@inf.ed.ac.uk>
parents:
diff changeset
11 import sys, argparse, regex, os, shutil, io, gzip, time, shlex
046dbe557911 write to tmp file implemented
Henry S. Thompson <ht@inf.ed.ac.uk>
parents:
diff changeset
12 from isal import igzip
046dbe557911 write to tmp file implemented
Henry S. Thompson <ht@inf.ed.ac.uk>
parents:
diff changeset
13 from subprocess import Popen, PIPE
046dbe557911 write to tmp file implemented
Henry S. Thompson <ht@inf.ed.ac.uk>
parents:
diff changeset
14 #import asyncio
046dbe557911 write to tmp file implemented
Henry S. Thompson <ht@inf.ed.ac.uk>
parents:
diff changeset
15
046dbe557911 write to tmp file implemented
Henry S. Thompson <ht@inf.ed.ac.uk>
parents:
diff changeset
16 HACK_USAGE=regex.compile('\[-x\]\n\s*\[length\] \[offset\] \[filename\]')
046dbe557911 write to tmp file implemented
Henry S. Thompson <ht@inf.ed.ac.uk>
parents:
diff changeset
17 BINOUT=sys.stdout.buffer
046dbe557911 write to tmp file implemented
Henry S. Thompson <ht@inf.ed.ac.uk>
parents:
diff changeset
18 FPAT="/%s/%s/orig/%s/%s"
046dbe557911 write to tmp file implemented
Henry S. Thompson <ht@inf.ed.ac.uk>
parents:
diff changeset
19
046dbe557911 write to tmp file implemented
Henry S. Thompson <ht@inf.ed.ac.uk>
parents:
diff changeset
20 CMD_PROC=None
046dbe557911 write to tmp file implemented
Henry S. Thompson <ht@inf.ed.ac.uk>
parents:
diff changeset
21 TMPFILENAME=None
046dbe557911 write to tmp file implemented
Henry S. Thompson <ht@inf.ed.ac.uk>
parents:
diff changeset
22
046dbe557911 write to tmp file implemented
Henry S. Thompson <ht@inf.ed.ac.uk>
parents:
diff changeset
23 class HackFormat(argparse.RawDescriptionHelpFormatter):
046dbe557911 write to tmp file implemented
Henry S. Thompson <ht@inf.ed.ac.uk>
parents:
diff changeset
24 def format_help(self):
234
36a9dc0d52d4 for debugging?
Henry S. Thompson <ht@inf.ed.ac.uk>
parents: 120
diff changeset
25 global FOO
18
046dbe557911 write to tmp file implemented
Henry S. Thompson <ht@inf.ed.ac.uk>
parents:
diff changeset
26 FOO=argparse.RawDescriptionHelpFormatter.format_help(self)
046dbe557911 write to tmp file implemented
Henry S. Thompson <ht@inf.ed.ac.uk>
parents:
diff changeset
27 return HACK_USAGE.sub('\n [ ( -x | length offset filename ) ]',
046dbe557911 write to tmp file implemented
Henry S. Thompson <ht@inf.ed.ac.uk>
parents:
diff changeset
28 FOO)
046dbe557911 write to tmp file implemented
Henry S. Thompson <ht@inf.ed.ac.uk>
parents:
diff changeset
29
046dbe557911 write to tmp file implemented
Henry S. Thompson <ht@inf.ed.ac.uk>
parents:
diff changeset
30 def process(options,buf,filename,offset,length,whole):
21
cbac7dfe2f24 interpolate process0, support permanent subproc
Henry S. Thompson <ht@inf.ed.ac.uk>
parents: 18
diff changeset
31 try:
cbac7dfe2f24 interpolate process0, support permanent subproc
Henry S. Thompson <ht@inf.ed.ac.uk>
parents: 18
diff changeset
32 process0(options,buf,filename,offset,length,whole)
cbac7dfe2f24 interpolate process0, support permanent subproc
Henry S. Thompson <ht@inf.ed.ac.uk>
parents: 18
diff changeset
33 except Exception as e:
cbac7dfe2f24 interpolate process0, support permanent subproc
Henry S. Thompson <ht@inf.ed.ac.uk>
parents: 18
diff changeset
34 if options.debug:
cbac7dfe2f24 interpolate process0, support permanent subproc
Henry S. Thompson <ht@inf.ed.ac.uk>
parents: 18
diff changeset
35 import traceback
cbac7dfe2f24 interpolate process0, support permanent subproc
Henry S. Thompson <ht@inf.ed.ac.uk>
parents: 18
diff changeset
36 traceback.print_exc(file=sys.stderr)
cbac7dfe2f24 interpolate process0, support permanent subproc
Henry S. Thompson <ht@inf.ed.ac.uk>
parents: 18
diff changeset
37 else:
238
a4538bdfa92a fix error message
Henry S. Thompson <ht@inf.ed.ac.uk>
parents: 234
diff changeset
38 print("Process fail: %s, input:\n length=%s, offset=%s, file=%s"%(e,
a4538bdfa92a fix error message
Henry S. Thompson <ht@inf.ed.ac.uk>
parents: 234
diff changeset
39 length,
a4538bdfa92a fix error message
Henry S. Thompson <ht@inf.ed.ac.uk>
parents: 234
diff changeset
40 offset,
a4538bdfa92a fix error message
Henry S. Thompson <ht@inf.ed.ac.uk>
parents: 234
diff changeset
41 filename),
a4538bdfa92a fix error message
Henry S. Thompson <ht@inf.ed.ac.uk>
parents: 234
diff changeset
42 file=sys.stderr)
21
cbac7dfe2f24 interpolate process0, support permanent subproc
Henry S. Thompson <ht@inf.ed.ac.uk>
parents: 18
diff changeset
43 exit(3)
cbac7dfe2f24 interpolate process0, support permanent subproc
Henry S. Thompson <ht@inf.ed.ac.uk>
parents: 18
diff changeset
44
cbac7dfe2f24 interpolate process0, support permanent subproc
Henry S. Thompson <ht@inf.ed.ac.uk>
parents: 18
diff changeset
45 def process0(options,buf,filename,offset,length,whole):
cbac7dfe2f24 interpolate process0, support permanent subproc
Henry S. Thompson <ht@inf.ed.ac.uk>
parents: 18
diff changeset
46 global TMPFILENAME, TMPFILE
18
046dbe557911 write to tmp file implemented
Henry S. Thompson <ht@inf.ed.ac.uk>
parents:
diff changeset
47 if options.save:
046dbe557911 write to tmp file implemented
Henry S. Thompson <ht@inf.ed.ac.uk>
parents:
diff changeset
48 (tf,TMPFILENAME)=tempfile.mkstemp()
046dbe557911 write to tmp file implemented
Henry S. Thompson <ht@inf.ed.ac.uk>
parents:
diff changeset
49 TMPFILE=open(tf,mode='wb')
21
cbac7dfe2f24 interpolate process0, support permanent subproc
Henry S. Thompson <ht@inf.ed.ac.uk>
parents: 18
diff changeset
50 if options.cmd and not options.process:
cbac7dfe2f24 interpolate process0, support permanent subproc
Henry S. Thompson <ht@inf.ed.ac.uk>
parents: 18
diff changeset
51 launch(options.cmd)
18
046dbe557911 write to tmp file implemented
Henry S. Thompson <ht@inf.ed.ac.uk>
parents:
diff changeset
52 process1(options,buf,filename,offset,length,whole)
046dbe557911 write to tmp file implemented
Henry S. Thompson <ht@inf.ed.ac.uk>
parents:
diff changeset
53 if options.save:
046dbe557911 write to tmp file implemented
Henry S. Thompson <ht@inf.ed.ac.uk>
parents:
diff changeset
54 TMPFILE.close()
046dbe557911 write to tmp file implemented
Henry S. Thompson <ht@inf.ed.ac.uk>
parents:
diff changeset
55 if options.cmd:
046dbe557911 write to tmp file implemented
Henry S. Thompson <ht@inf.ed.ac.uk>
parents:
diff changeset
56 _output_subproc(bytes(TMPFILENAME,'utf-8'))
046dbe557911 write to tmp file implemented
Henry S. Thompson <ht@inf.ed.ac.uk>
parents:
diff changeset
57 _output_subproc(b"\n")
046dbe557911 write to tmp file implemented
Henry S. Thompson <ht@inf.ed.ac.uk>
parents:
diff changeset
58 else:
046dbe557911 write to tmp file implemented
Henry S. Thompson <ht@inf.ed.ac.uk>
parents:
diff changeset
59 BINOUT.write(bytes(TMPFILENAME,'utf-8'))
046dbe557911 write to tmp file implemented
Henry S. Thompson <ht@inf.ed.ac.uk>
parents:
diff changeset
60 BINOUT.write(b"\n")
046dbe557911 write to tmp file implemented
Henry S. Thompson <ht@inf.ed.ac.uk>
parents:
diff changeset
61 if options.cmd:
21
cbac7dfe2f24 interpolate process0, support permanent subproc
Henry S. Thompson <ht@inf.ed.ac.uk>
parents: 18
diff changeset
62 if not options.process:
cbac7dfe2f24 interpolate process0, support permanent subproc
Henry S. Thompson <ht@inf.ed.ac.uk>
parents: 18
diff changeset
63 windup(filename,options,length)
18
046dbe557911 write to tmp file implemented
Henry S. Thompson <ht@inf.ed.ac.uk>
parents:
diff changeset
64 if options.save:
046dbe557911 write to tmp file implemented
Henry S. Thompson <ht@inf.ed.ac.uk>
parents:
diff changeset
65 os.unlink(TMPFILENAME)
046dbe557911 write to tmp file implemented
Henry S. Thompson <ht@inf.ed.ac.uk>
parents:
diff changeset
66 TMPFILENAME=None
046dbe557911 write to tmp file implemented
Henry S. Thompson <ht@inf.ed.ac.uk>
parents:
diff changeset
67 elif options.save:
046dbe557911 write to tmp file implemented
Henry S. Thompson <ht@inf.ed.ac.uk>
parents:
diff changeset
68 print("%s will need to be deleted"%TMPFILENAME,file=sys.stderr)
046dbe557911 write to tmp file implemented
Henry S. Thompson <ht@inf.ed.ac.uk>
parents:
diff changeset
69 TMPFILENAME=None
046dbe557911 write to tmp file implemented
Henry S. Thompson <ht@inf.ed.ac.uk>
parents:
diff changeset
70
21
cbac7dfe2f24 interpolate process0, support permanent subproc
Henry S. Thompson <ht@inf.ed.ac.uk>
parents: 18
diff changeset
71 def launch(cmd):
cbac7dfe2f24 interpolate process0, support permanent subproc
Henry S. Thompson <ht@inf.ed.ac.uk>
parents: 18
diff changeset
72 global CMD_PROC, BINOUT
cbac7dfe2f24 interpolate process0, support permanent subproc
Henry S. Thompson <ht@inf.ed.ac.uk>
parents: 18
diff changeset
73 CMD_PROC=Popen(shlex.split(cmd),stdin=PIPE,bufsize=0)
cbac7dfe2f24 interpolate process0, support permanent subproc
Henry S. Thompson <ht@inf.ed.ac.uk>
parents: 18
diff changeset
74 BINOUT=CMD_PROC.stdin
cbac7dfe2f24 interpolate process0, support permanent subproc
Henry S. Thompson <ht@inf.ed.ac.uk>
parents: 18
diff changeset
75
cbac7dfe2f24 interpolate process0, support permanent subproc
Henry S. Thompson <ht@inf.ed.ac.uk>
parents: 18
diff changeset
76 def windup(length,offset,filename):
cbac7dfe2f24 interpolate process0, support permanent subproc
Henry S. Thompson <ht@inf.ed.ac.uk>
parents: 18
diff changeset
77 # Wind up subproc
cbac7dfe2f24 interpolate process0, support permanent subproc
Henry S. Thompson <ht@inf.ed.ac.uk>
parents: 18
diff changeset
78 BINOUT.close()
cbac7dfe2f24 interpolate process0, support permanent subproc
Henry S. Thompson <ht@inf.ed.ac.uk>
parents: 18
diff changeset
79 if CMD_PROC.wait()!=0: # could/should be async?
cbac7dfe2f24 interpolate process0, support permanent subproc
Henry S. Thompson <ht@inf.ed.ac.uk>
parents: 18
diff changeset
80 print("subproc of %s:%s:%s failed with %s"%(length,offset,filename,
cbac7dfe2f24 interpolate process0, support permanent subproc
Henry S. Thompson <ht@inf.ed.ac.uk>
parents: 18
diff changeset
81 CMD_PROC.returncode),
cbac7dfe2f24 interpolate process0, support permanent subproc
Henry S. Thompson <ht@inf.ed.ac.uk>
parents: 18
diff changeset
82 file=sys.stderr)
cbac7dfe2f24 interpolate process0, support permanent subproc
Henry S. Thompson <ht@inf.ed.ac.uk>
parents: 18
diff changeset
83
18
046dbe557911 write to tmp file implemented
Henry S. Thompson <ht@inf.ed.ac.uk>
parents:
diff changeset
84 def _output_tmpfile(buf):
046dbe557911 write to tmp file implemented
Henry S. Thompson <ht@inf.ed.ac.uk>
parents:
diff changeset
85 TMPFILE.write(buf)
046dbe557911 write to tmp file implemented
Henry S. Thompson <ht@inf.ed.ac.uk>
parents:
diff changeset
86
046dbe557911 write to tmp file implemented
Henry S. Thompson <ht@inf.ed.ac.uk>
parents:
diff changeset
87 def _output_stdout(buf):
046dbe557911 write to tmp file implemented
Henry S. Thompson <ht@inf.ed.ac.uk>
parents:
diff changeset
88 BINOUT.write(buf)
046dbe557911 write to tmp file implemented
Henry S. Thompson <ht@inf.ed.ac.uk>
parents:
diff changeset
89
046dbe557911 write to tmp file implemented
Henry S. Thompson <ht@inf.ed.ac.uk>
parents:
diff changeset
90 def _output_subproc(buf):
046dbe557911 write to tmp file implemented
Henry S. Thompson <ht@inf.ed.ac.uk>
parents:
diff changeset
91 toWrite=len(buf)
046dbe557911 write to tmp file implemented
Henry S. Thompson <ht@inf.ed.ac.uk>
parents:
diff changeset
92 while toWrite>0:
046dbe557911 write to tmp file implemented
Henry S. Thompson <ht@inf.ed.ac.uk>
parents:
diff changeset
93 toWrite -= BINOUT.write(buf)
046dbe557911 write to tmp file implemented
Henry S. Thompson <ht@inf.ed.ac.uk>
parents:
diff changeset
94
046dbe557911 write to tmp file implemented
Henry S. Thompson <ht@inf.ed.ac.uk>
parents:
diff changeset
95 def process1(options,buf,filename,offset,length,whole):
046dbe557911 write to tmp file implemented
Henry S. Thompson <ht@inf.ed.ac.uk>
parents:
diff changeset
96 root=options.root
046dbe557911 write to tmp file implemented
Henry S. Thompson <ht@inf.ed.ac.uk>
parents:
diff changeset
97 rfn=root+filename
046dbe557911 write to tmp file implemented
Henry S. Thompson <ht@inf.ed.ac.uk>
parents:
diff changeset
98 if root!="/beegfs/common_crawl":
046dbe557911 write to tmp file implemented
Henry S. Thompson <ht@inf.ed.ac.uk>
parents:
diff changeset
99 # Support using ramdisk or other local disk as a faster cached
046dbe557911 write to tmp file implemented
Henry S. Thompson <ht@inf.ed.ac.uk>
parents:
diff changeset
100 if not os.path.exists(rfn):
046dbe557911 write to tmp file implemented
Henry S. Thompson <ht@inf.ed.ac.uk>
parents:
diff changeset
101 if not os.path.exists(os.path.dirname(rfn)):
046dbe557911 write to tmp file implemented
Henry S. Thompson <ht@inf.ed.ac.uk>
parents:
diff changeset
102 os.makedirs(os.path.dirname(rfn))
046dbe557911 write to tmp file implemented
Henry S. Thompson <ht@inf.ed.ac.uk>
parents:
diff changeset
103 with io.FileIO('/beegfs/common_crawl'+filename,'r') as infile, \
046dbe557911 write to tmp file implemented
Henry S. Thompson <ht@inf.ed.ac.uk>
parents:
diff changeset
104 io.FileIO(rfn,'w') as outfile:
046dbe557911 write to tmp file implemented
Henry S. Thompson <ht@inf.ed.ac.uk>
parents:
diff changeset
105 #shutil.copyfileobj(infile,outfile,128*1024*1024)
046dbe557911 write to tmp file implemented
Henry S. Thompson <ht@inf.ed.ac.uk>
parents:
diff changeset
106 while True:
046dbe557911 write to tmp file implemented
Henry S. Thompson <ht@inf.ed.ac.uk>
parents:
diff changeset
107 l=infile.readinto(buf)
046dbe557911 write to tmp file implemented
Henry S. Thompson <ht@inf.ed.ac.uk>
parents:
diff changeset
108 if l==0:
046dbe557911 write to tmp file implemented
Henry S. Thompson <ht@inf.ed.ac.uk>
parents:
diff changeset
109 break
046dbe557911 write to tmp file implemented
Henry S. Thompson <ht@inf.ed.ac.uk>
parents:
diff changeset
110 outfile.write(memoryview(buf)[:l])
046dbe557911 write to tmp file implemented
Henry S. Thompson <ht@inf.ed.ac.uk>
parents:
diff changeset
111 file=open(rfn,'rb',0)
046dbe557911 write to tmp file implemented
Henry S. Thompson <ht@inf.ed.ac.uk>
parents:
diff changeset
112 file.seek(offset)
046dbe557911 write to tmp file implemented
Henry S. Thompson <ht@inf.ed.ac.uk>
parents:
diff changeset
113 bv=memoryview(buf)[:length]
046dbe557911 write to tmp file implemented
Henry S. Thompson <ht@inf.ed.ac.uk>
parents:
diff changeset
114 nb=file.readinto(bv)
046dbe557911 write to tmp file implemented
Henry S. Thompson <ht@inf.ed.ac.uk>
parents:
diff changeset
115 file.close()
046dbe557911 write to tmp file implemented
Henry S. Thompson <ht@inf.ed.ac.uk>
parents:
diff changeset
116 if nb!=length:
046dbe557911 write to tmp file implemented
Henry S. Thompson <ht@inf.ed.ac.uk>
parents:
diff changeset
117 raise ValueError("Chunk read losing: %s, got %s expected %s at %s"%(file.name,
046dbe557911 write to tmp file implemented
Henry S. Thompson <ht@inf.ed.ac.uk>
parents:
diff changeset
118 nb,length,offset))
046dbe557911 write to tmp file implemented
Henry S. Thompson <ht@inf.ed.ac.uk>
parents:
diff changeset
119 if whole and options.zipped:
046dbe557911 write to tmp file implemented
Henry S. Thompson <ht@inf.ed.ac.uk>
parents:
diff changeset
120 _output(bv)
046dbe557911 write to tmp file implemented
Henry S. Thompson <ht@inf.ed.ac.uk>
parents:
diff changeset
121 return
046dbe557911 write to tmp file implemented
Henry S. Thompson <ht@inf.ed.ac.uk>
parents:
diff changeset
122 gzip_chunk = io.BytesIO(bv)
046dbe557911 write to tmp file implemented
Henry S. Thompson <ht@inf.ed.ac.uk>
parents:
diff changeset
123 uv=memoryview(buf)[length:]
046dbe557911 write to tmp file implemented
Henry S. Thompson <ht@inf.ed.ac.uk>
parents:
diff changeset
124 with igzip.IGzipFile(fileobj=gzip_chunk) as gzip_fin:
046dbe557911 write to tmp file implemented
Henry S. Thompson <ht@inf.ed.ac.uk>
parents:
diff changeset
125 ll=0
046dbe557911 write to tmp file implemented
Henry S. Thompson <ht@inf.ed.ac.uk>
parents:
diff changeset
126 while True:
046dbe557911 write to tmp file implemented
Henry S. Thompson <ht@inf.ed.ac.uk>
parents:
diff changeset
127 l=gzip_fin.readinto(uv)
046dbe557911 write to tmp file implemented
Henry S. Thompson <ht@inf.ed.ac.uk>
parents:
diff changeset
128 if not l:
046dbe557911 write to tmp file implemented
Henry S. Thompson <ht@inf.ed.ac.uk>
parents:
diff changeset
129 break
046dbe557911 write to tmp file implemented
Henry S. Thompson <ht@inf.ed.ac.uk>
parents:
diff changeset
130 ll+=l
046dbe557911 write to tmp file implemented
Henry S. Thompson <ht@inf.ed.ac.uk>
parents:
diff changeset
131 cb=memoryview(uv)[:ll]
046dbe557911 write to tmp file implemented
Henry S. Thompson <ht@inf.ed.ac.uk>
parents:
diff changeset
132 if whole:
046dbe557911 write to tmp file implemented
Henry S. Thompson <ht@inf.ed.ac.uk>
parents:
diff changeset
133 _output(cb)
046dbe557911 write to tmp file implemented
Henry S. Thompson <ht@inf.ed.ac.uk>
parents:
diff changeset
134 return
046dbe557911 write to tmp file implemented
Henry S. Thompson <ht@inf.ed.ac.uk>
parents:
diff changeset
135 # Only output parts (0 = WARC header, 1 = HTTP header, 2 = body) that are wanted
046dbe557911 write to tmp file implemented
Henry S. Thompson <ht@inf.ed.ac.uk>
parents:
diff changeset
136 state=0
046dbe557911 write to tmp file implemented
Henry S. Thompson <ht@inf.ed.ac.uk>
parents:
diff changeset
137 tr=None # Was this record truncated?
046dbe557911 write to tmp file implemented
Henry S. Thompson <ht@inf.ed.ac.uk>
parents:
diff changeset
138 bl=None # for HTTP Content-Length for the length of the body?
21
cbac7dfe2f24 interpolate process0, support permanent subproc
Henry S. Thompson <ht@inf.ed.ac.uk>
parents: 18
diff changeset
139 # Could we make this faster by working purely within the cb memoryview?
cbac7dfe2f24 interpolate process0, support permanent subproc
Henry S. Thompson <ht@inf.ed.ac.uk>
parents: 18
diff changeset
140 # It would be messy, but avoid copying huge amounts
cbac7dfe2f24 interpolate process0, support permanent subproc
Henry S. Thompson <ht@inf.ed.ac.uk>
parents: 18
diff changeset
141 # The outer loop would just be something like
cbac7dfe2f24 interpolate process0, support permanent subproc
Henry S. Thompson <ht@inf.ed.ac.uk>
parents: 18
diff changeset
142 # clbv=memoryview(bytearray(b"Content-Length: "))
cbac7dfe2f24 interpolate process0, support permanent subproc
Henry S. Thompson <ht@inf.ed.ac.uk>
parents: 18
diff changeset
143 # i=s=0
cbac7dfe2f24 interpolate process0, support permanent subproc
Henry S. Thompson <ht@inf.ed.ac.uk>
parents: 18
diff changeset
144 # while i<ll:
cbac7dfe2f24 interpolate process0, support permanent subproc
Henry S. Thompson <ht@inf.ed.ac.uk>
parents: 18
diff changeset
145 # if cb[i]==10: # need to handle \r\n
cbac7dfe2f24 interpolate process0, support permanent subproc
Henry S. Thompson <ht@inf.ed.ac.uk>
parents: 18
diff changeset
146 # L=cb[s:i]
cbac7dfe2f24 interpolate process0, support permanent subproc
Henry S. Thompson <ht@inf.ed.ac.uk>
parents: 18
diff changeset
147 # s=i=i+1
cbac7dfe2f24 interpolate process0, support permanent subproc
Henry S. Thompson <ht@inf.ed.ac.uk>
parents: 18
diff changeset
148 # if L[:16]==clbv:
cbac7dfe2f24 interpolate process0, support permanent subproc
Henry S. Thompson <ht@inf.ed.ac.uk>
parents: 18
diff changeset
149 # wl=int(L[16:])
cbac7dfe2f24 interpolate process0, support permanent subproc
Henry S. Thompson <ht@inf.ed.ac.uk>
parents: 18
diff changeset
150 # else:
cbac7dfe2f24 interpolate process0, support permanent subproc
Henry S. Thompson <ht@inf.ed.ac.uk>
parents: 18
diff changeset
151 # i+=1
cbac7dfe2f24 interpolate process0, support permanent subproc
Henry S. Thompson <ht@inf.ed.ac.uk>
parents: 18
diff changeset
152 #
18
046dbe557911 write to tmp file implemented
Henry S. Thompson <ht@inf.ed.ac.uk>
parents:
diff changeset
153 with io.BytesIO(cb) as clear_text:
046dbe557911 write to tmp file implemented
Henry S. Thompson <ht@inf.ed.ac.uk>
parents:
diff changeset
154 for L in clear_text:
046dbe557911 write to tmp file implemented
Henry S. Thompson <ht@inf.ed.ac.uk>
parents:
diff changeset
155 if state==0:
046dbe557911 write to tmp file implemented
Henry S. Thompson <ht@inf.ed.ac.uk>
parents:
diff changeset
156 # WARC header
046dbe557911 write to tmp file implemented
Henry S. Thompson <ht@inf.ed.ac.uk>
parents:
diff changeset
157 if L.startswith(b"Content-Length: "):
046dbe557911 write to tmp file implemented
Henry S. Thompson <ht@inf.ed.ac.uk>
parents:
diff changeset
158 wl=int(L[16:].rstrip())
046dbe557911 write to tmp file implemented
Henry S. Thompson <ht@inf.ed.ac.uk>
parents:
diff changeset
159 elif L.startswith(b"WARC-Truncated: "):
046dbe557911 write to tmp file implemented
Henry S. Thompson <ht@inf.ed.ac.uk>
parents:
diff changeset
160 tr=L[16:].rstrip()
046dbe557911 write to tmp file implemented
Henry S. Thompson <ht@inf.ed.ac.uk>
parents:
diff changeset
161 tr="EMPTY" if tr=="" else tr
046dbe557911 write to tmp file implemented
Henry S. Thompson <ht@inf.ed.ac.uk>
parents:
diff changeset
162 elif L==b"" or L.startswith(b"\r"): # for idempotency
046dbe557911 write to tmp file implemented
Henry S. Thompson <ht@inf.ed.ac.uk>
parents:
diff changeset
163 # Blank line, WARC header is finished
046dbe557911 write to tmp file implemented
Henry S. Thompson <ht@inf.ed.ac.uk>
parents:
diff changeset
164 if not (options.headers or options.body):
046dbe557911 write to tmp file implemented
Henry S. Thompson <ht@inf.ed.ac.uk>
parents:
diff changeset
165 return
046dbe557911 write to tmp file implemented
Henry S. Thompson <ht@inf.ed.ac.uk>
parents:
diff changeset
166 state=1
046dbe557911 write to tmp file implemented
Henry S. Thompson <ht@inf.ed.ac.uk>
parents:
diff changeset
167 # Note we preserve the empty line
046dbe557911 write to tmp file implemented
Henry S. Thompson <ht@inf.ed.ac.uk>
parents:
diff changeset
168 if options.warc:
046dbe557911 write to tmp file implemented
Henry S. Thompson <ht@inf.ed.ac.uk>
parents:
diff changeset
169 _output(L)
046dbe557911 write to tmp file implemented
Henry S. Thompson <ht@inf.ed.ac.uk>
parents:
diff changeset
170 continue
046dbe557911 write to tmp file implemented
Henry S. Thompson <ht@inf.ed.ac.uk>
parents:
diff changeset
171 if state==1:
046dbe557911 write to tmp file implemented
Henry S. Thompson <ht@inf.ed.ac.uk>
parents:
diff changeset
172 # HTTP header
046dbe557911 write to tmp file implemented
Henry S. Thompson <ht@inf.ed.ac.uk>
parents:
diff changeset
173 wl -= len(L)
046dbe557911 write to tmp file implemented
Henry S. Thompson <ht@inf.ed.ac.uk>
parents:
diff changeset
174 if not (L==b"" or L.startswith(b"\r")):
046dbe557911 write to tmp file implemented
Henry S. Thompson <ht@inf.ed.ac.uk>
parents:
diff changeset
175 # Non-blank, it's a header
21
cbac7dfe2f24 interpolate process0, support permanent subproc
Henry S. Thompson <ht@inf.ed.ac.uk>
parents: 18
diff changeset
176 (h,_,v)=L.partition(b": ")
cbac7dfe2f24 interpolate process0, support permanent subproc
Henry S. Thompson <ht@inf.ed.ac.uk>
parents: 18
diff changeset
177 if bl is None and (h==b"Content-Length"):
cbac7dfe2f24 interpolate process0, support permanent subproc
Henry S. Thompson <ht@inf.ed.ac.uk>
parents: 18
diff changeset
178 bl=int(v)
18
046dbe557911 write to tmp file implemented
Henry S. Thompson <ht@inf.ed.ac.uk>
parents:
diff changeset
179 if options.headers:
21
cbac7dfe2f24 interpolate process0, support permanent subproc
Henry S. Thompson <ht@inf.ed.ac.uk>
parents: 18
diff changeset
180 if isinstance(options.headers,dict):
cbac7dfe2f24 interpolate process0, support permanent subproc
Henry S. Thompson <ht@inf.ed.ac.uk>
parents: 18
diff changeset
181 if h in options.headers:
cbac7dfe2f24 interpolate process0, support permanent subproc
Henry S. Thompson <ht@inf.ed.ac.uk>
parents: 18
diff changeset
182 options.headers[h]=v
cbac7dfe2f24 interpolate process0, support permanent subproc
Henry S. Thompson <ht@inf.ed.ac.uk>
parents: 18
diff changeset
183 else:
cbac7dfe2f24 interpolate process0, support permanent subproc
Henry S. Thompson <ht@inf.ed.ac.uk>
parents: 18
diff changeset
184 _output(L)
18
046dbe557911 write to tmp file implemented
Henry S. Thompson <ht@inf.ed.ac.uk>
parents:
diff changeset
185 else:
046dbe557911 write to tmp file implemented
Henry S. Thompson <ht@inf.ed.ac.uk>
parents:
diff changeset
186 # Blank line, HTTP header is finished
21
cbac7dfe2f24 interpolate process0, support permanent subproc
Henry S. Thompson <ht@inf.ed.ac.uk>
parents: 18
diff changeset
187 if isinstance(options.headers,dict):
cbac7dfe2f24 interpolate process0, support permanent subproc
Henry S. Thompson <ht@inf.ed.ac.uk>
parents: 18
diff changeset
188 _output(bytes(str(options.headers),'utf-8'))
18
046dbe557911 write to tmp file implemented
Henry S. Thompson <ht@inf.ed.ac.uk>
parents:
diff changeset
189 if not options.body:
046dbe557911 write to tmp file implemented
Henry S. Thompson <ht@inf.ed.ac.uk>
parents:
diff changeset
190 return
046dbe557911 write to tmp file implemented
Henry S. Thompson <ht@inf.ed.ac.uk>
parents:
diff changeset
191 if options.headers:
046dbe557911 write to tmp file implemented
Henry S. Thompson <ht@inf.ed.ac.uk>
parents:
diff changeset
192 _output(L)
046dbe557911 write to tmp file implemented
Henry S. Thompson <ht@inf.ed.ac.uk>
parents:
diff changeset
193 state=2
046dbe557911 write to tmp file implemented
Henry S. Thompson <ht@inf.ed.ac.uk>
parents:
diff changeset
194 # The above is just for sanity, because we do _not_
046dbe557911 write to tmp file implemented
Henry S. Thompson <ht@inf.ed.ac.uk>
parents:
diff changeset
195 # continue with the outer loop,
046dbe557911 write to tmp file implemented
Henry S. Thompson <ht@inf.ed.ac.uk>
parents:
diff changeset
196 # since we can now block-output the entire rest of the
046dbe557911 write to tmp file implemented
Henry S. Thompson <ht@inf.ed.ac.uk>
parents:
diff changeset
197 # input buffer.
046dbe557911 write to tmp file implemented
Henry S. Thompson <ht@inf.ed.ac.uk>
parents:
diff changeset
198 if bl is not None:
046dbe557911 write to tmp file implemented
Henry S. Thompson <ht@inf.ed.ac.uk>
parents:
diff changeset
199 if bl!=wl:
046dbe557911 write to tmp file implemented
Henry S. Thompson <ht@inf.ed.ac.uk>
parents:
diff changeset
200 print("length mismatch: %s %s %s here: %s given: %s trunc: %s"%\
046dbe557911 write to tmp file implemented
Henry S. Thompson <ht@inf.ed.ac.uk>
parents:
diff changeset
201 (length,offset,filename,wl,bl,tr),file=sys.stderr)
046dbe557911 write to tmp file implemented
Henry S. Thompson <ht@inf.ed.ac.uk>
parents:
diff changeset
202 # HTTP body
046dbe557911 write to tmp file implemented
Henry S. Thompson <ht@inf.ed.ac.uk>
parents:
diff changeset
203 balance=clear_text.tell()
046dbe557911 write to tmp file implemented
Henry S. Thompson <ht@inf.ed.ac.uk>
parents:
diff changeset
204 #print(balance,bl,wl,ll,ll-balance,file=sys.stderr)
046dbe557911 write to tmp file implemented
Henry S. Thompson <ht@inf.ed.ac.uk>
parents:
diff changeset
205 # Output whatever is left
046dbe557911 write to tmp file implemented
Henry S. Thompson <ht@inf.ed.ac.uk>
parents:
diff changeset
206 _output(cb[balance:balance+wl])
046dbe557911 write to tmp file implemented
Henry S. Thompson <ht@inf.ed.ac.uk>
parents:
diff changeset
207 return
046dbe557911 write to tmp file implemented
Henry S. Thompson <ht@inf.ed.ac.uk>
parents:
diff changeset
208
046dbe557911 write to tmp file implemented
Henry S. Thompson <ht@inf.ed.ac.uk>
parents:
diff changeset
209 def main():
046dbe557911 write to tmp file implemented
Henry S. Thompson <ht@inf.ed.ac.uk>
parents:
diff changeset
210 global _output,TMPFILE,TMPFILENAME,tempfile
046dbe557911 write to tmp file implemented
Henry S. Thompson <ht@inf.ed.ac.uk>
parents:
diff changeset
211 parser = argparse.ArgumentParser(
046dbe557911 write to tmp file implemented
Henry S. Thompson <ht@inf.ed.ac.uk>
parents:
diff changeset
212 description='''Extract records from warc files given length, offset and file triples.
046dbe557911 write to tmp file implemented
Henry S. Thompson <ht@inf.ed.ac.uk>
parents:
diff changeset
213 Input one triple on command line, or
046dbe557911 write to tmp file implemented
Henry S. Thompson <ht@inf.ed.ac.uk>
parents:
diff changeset
214 triples from stdin as tab-delimited lines
046dbe557911 write to tmp file implemented
Henry S. Thompson <ht@inf.ed.ac.uk>
parents:
diff changeset
215 or complete cdx index lines.
046dbe557911 write to tmp file implemented
Henry S. Thompson <ht@inf.ed.ac.uk>
parents:
diff changeset
216 In all cases by 'filename' is meant crawlid/segmentid/type/filename''',
046dbe557911 write to tmp file implemented
Henry S. Thompson <ht@inf.ed.ac.uk>
parents:
diff changeset
217 epilog='''Note that if no output flag(s) is/are given,
046dbe557911 write to tmp file implemented
Henry S. Thompson <ht@inf.ed.ac.uk>
parents:
diff changeset
218 the whole WARC record will be output, more efficiently than
046dbe557911 write to tmp file implemented
Henry S. Thompson <ht@inf.ed.ac.uk>
parents:
diff changeset
219 would be the case if all three flags were given.''',
046dbe557911 write to tmp file implemented
Henry S. Thompson <ht@inf.ed.ac.uk>
parents:
diff changeset
220 add_help=False,
046dbe557911 write to tmp file implemented
Henry S. Thompson <ht@inf.ed.ac.uk>
parents:
diff changeset
221 conflict_handler='resolve',
046dbe557911 write to tmp file implemented
Henry S. Thompson <ht@inf.ed.ac.uk>
parents:
diff changeset
222 formatter_class=HackFormat
046dbe557911 write to tmp file implemented
Henry S. Thompson <ht@inf.ed.ac.uk>
parents:
diff changeset
223 )
046dbe557911 write to tmp file implemented
Henry S. Thompson <ht@inf.ed.ac.uk>
parents:
diff changeset
224 fphelp=('format string for turning 4 filename components into a path, must contain %%s exactly 4 times,\ndefault is "%s"'%FPAT).replace('%s','%%s')
046dbe557911 write to tmp file implemented
Henry S. Thompson <ht@inf.ed.ac.uk>
parents:
diff changeset
225 parser.add_argument('--help',help='Show help',action='help')
046dbe557911 write to tmp file implemented
Henry S. Thompson <ht@inf.ed.ac.uk>
parents:
diff changeset
226 parser.add_argument('-d','--debug',help='Debug output',action='store_true')
046dbe557911 write to tmp file implemented
Henry S. Thompson <ht@inf.ed.ac.uk>
parents:
diff changeset
227 parser.add_argument('-w','--warc',help='output WARC headers',
046dbe557911 write to tmp file implemented
Henry S. Thompson <ht@inf.ed.ac.uk>
parents:
diff changeset
228 action='store_true')
21
cbac7dfe2f24 interpolate process0, support permanent subproc
Henry S. Thompson <ht@inf.ed.ac.uk>
parents: 18
diff changeset
229 parser.add_argument('-h','--headers',help='process HTTP headers: collect into dict with named values (,-separated) if arg present, else output',
cbac7dfe2f24 interpolate process0, support permanent subproc
Henry S. Thompson <ht@inf.ed.ac.uk>
parents: 18
diff changeset
230 nargs='?',default=None,const=True)
18
046dbe557911 write to tmp file implemented
Henry S. Thompson <ht@inf.ed.ac.uk>
parents:
diff changeset
231 parser.add_argument('-b','--body',help='output HTTP body',
046dbe557911 write to tmp file implemented
Henry S. Thompson <ht@inf.ed.ac.uk>
parents:
diff changeset
232 action='store_true')
046dbe557911 write to tmp file implemented
Henry S. Thompson <ht@inf.ed.ac.uk>
parents:
diff changeset
233 parser.add_argument('-c','--cmd',help='pipes each result thru CMD')
21
cbac7dfe2f24 interpolate process0, support permanent subproc
Henry S. Thompson <ht@inf.ed.ac.uk>
parents: 18
diff changeset
234 parser.add_argument('-p','--process',help='with -c, launches CMD only once',
cbac7dfe2f24 interpolate process0, support permanent subproc
Henry S. Thompson <ht@inf.ed.ac.uk>
parents: 18
diff changeset
235 action='store_true')
18
046dbe557911 write to tmp file implemented
Henry S. Thompson <ht@inf.ed.ac.uk>
parents:
diff changeset
236 parser.add_argument('-m','--module.function',help='module.function to call with a stream'),
046dbe557911 write to tmp file implemented
Henry S. Thompson <ht@inf.ed.ac.uk>
parents:
diff changeset
237 parser.add_argument('-s','--save',action='store_true',
046dbe557911 write to tmp file implemented
Henry S. Thompson <ht@inf.ed.ac.uk>
parents:
diff changeset
238 help="write to a temporary file and output the name")
046dbe557911 write to tmp file implemented
Henry S. Thompson <ht@inf.ed.ac.uk>
parents:
diff changeset
239 parser.add_argument('-f','--fpath',
046dbe557911 write to tmp file implemented
Henry S. Thompson <ht@inf.ed.ac.uk>
parents:
diff changeset
240 help=fphelp,
046dbe557911 write to tmp file implemented
Henry S. Thompson <ht@inf.ed.ac.uk>
parents:
diff changeset
241 default=FPAT)
046dbe557911 write to tmp file implemented
Henry S. Thompson <ht@inf.ed.ac.uk>
parents:
diff changeset
242 parser.add_argument('-r','--root',nargs='?',
046dbe557911 write to tmp file implemented
Henry S. Thompson <ht@inf.ed.ac.uk>
parents:
diff changeset
243 help='File path root, create a copy there if necessary',
046dbe557911 write to tmp file implemented
Henry S. Thompson <ht@inf.ed.ac.uk>
parents:
diff changeset
244 default='/beegfs/common_crawl'),
046dbe557911 write to tmp file implemented
Henry S. Thompson <ht@inf.ed.ac.uk>
parents:
diff changeset
245 parser.add_argument('-z','--zipped',
046dbe557911 write to tmp file implemented
Henry S. Thompson <ht@inf.ed.ac.uk>
parents:
diff changeset
246 help="output raw gzipped record, ignored if any of -bhw supplied",
046dbe557911 write to tmp file implemented
Henry S. Thompson <ht@inf.ed.ac.uk>
parents:
diff changeset
247 action='store_true')
046dbe557911 write to tmp file implemented
Henry S. Thompson <ht@inf.ed.ac.uk>
parents:
diff changeset
248 sg=parser.add_mutually_exclusive_group()
046dbe557911 write to tmp file implemented
Henry S. Thompson <ht@inf.ed.ac.uk>
parents:
diff changeset
249 sg.add_argument('-x','--index',
046dbe557911 write to tmp file implemented
Henry S. Thompson <ht@inf.ed.ac.uk>
parents:
diff changeset
250 help='take lines of triples from a cdx index file as input',
046dbe557911 write to tmp file implemented
Henry S. Thompson <ht@inf.ed.ac.uk>
parents:
diff changeset
251 action='store_true')
046dbe557911 write to tmp file implemented
Henry S. Thompson <ht@inf.ed.ac.uk>
parents:
diff changeset
252 sg.add_argument('length',type=int,
046dbe557911 write to tmp file implemented
Henry S. Thompson <ht@inf.ed.ac.uk>
parents:
diff changeset
253 help='length in bytes of gzipped record',
046dbe557911 write to tmp file implemented
Henry S. Thompson <ht@inf.ed.ac.uk>
parents:
diff changeset
254 nargs='?')
046dbe557911 write to tmp file implemented
Henry S. Thompson <ht@inf.ed.ac.uk>
parents:
diff changeset
255 parser.add_argument('offset',type=int,
046dbe557911 write to tmp file implemented
Henry S. Thompson <ht@inf.ed.ac.uk>
parents:
diff changeset
256 help='start position in bytes of gzipped record',
046dbe557911 write to tmp file implemented
Henry S. Thompson <ht@inf.ed.ac.uk>
parents:
diff changeset
257 nargs='?')
046dbe557911 write to tmp file implemented
Henry S. Thompson <ht@inf.ed.ac.uk>
parents:
diff changeset
258 parser.add_argument('filename',
046dbe557911 write to tmp file implemented
Henry S. Thompson <ht@inf.ed.ac.uk>
parents:
diff changeset
259 help='pathname of gzipped Common Crawl WARC-format file',
046dbe557911 write to tmp file implemented
Henry S. Thompson <ht@inf.ed.ac.uk>
parents:
diff changeset
260 nargs='?')
046dbe557911 write to tmp file implemented
Henry S. Thompson <ht@inf.ed.ac.uk>
parents:
diff changeset
261 # Hack the order of optional and positional in the help output
046dbe557911 write to tmp file implemented
Henry S. Thompson <ht@inf.ed.ac.uk>
parents:
diff changeset
262 parser._action_groups.sort(key=lambda g:g.title)
046dbe557911 write to tmp file implemented
Henry S. Thompson <ht@inf.ed.ac.uk>
parents:
diff changeset
263 #parser.print_help()
046dbe557911 write to tmp file implemented
Henry S. Thompson <ht@inf.ed.ac.uk>
parents:
diff changeset
264 pa=parser.parse_args(sys.argv[1:])
046dbe557911 write to tmp file implemented
Henry S. Thompson <ht@inf.ed.ac.uk>
parents:
diff changeset
265 #print(pa,file=sys.stderr)
046dbe557911 write to tmp file implemented
Henry S. Thompson <ht@inf.ed.ac.uk>
parents:
diff changeset
266 if pa.length is not None:
046dbe557911 write to tmp file implemented
Henry S. Thompson <ht@inf.ed.ac.uk>
parents:
diff changeset
267 # We have to enforce our own check..
046dbe557911 write to tmp file implemented
Henry S. Thompson <ht@inf.ed.ac.uk>
parents:
diff changeset
268 if pa.offset is None or pa.filename is None:
046dbe557911 write to tmp file implemented
Henry S. Thompson <ht@inf.ed.ac.uk>
parents:
diff changeset
269 parser.error("length, offset and filename must all be supplied together")
21
cbac7dfe2f24 interpolate process0, support permanent subproc
Henry S. Thompson <ht@inf.ed.ac.uk>
parents: 18
diff changeset
270 if isinstance(pa.headers,str):
cbac7dfe2f24 interpolate process0, support permanent subproc
Henry S. Thompson <ht@inf.ed.ac.uk>
parents: 18
diff changeset
271 pa.headers=dict((bytes(k,'utf-8'),None) for k in pa.headers.split(','))
18
046dbe557911 write to tmp file implemented
Henry S. Thompson <ht@inf.ed.ac.uk>
parents:
diff changeset
272
046dbe557911 write to tmp file implemented
Henry S. Thompson <ht@inf.ed.ac.uk>
parents:
diff changeset
273 buf=bytearray(128*1024*1024)
046dbe557911 write to tmp file implemented
Henry S. Thompson <ht@inf.ed.ac.uk>
parents:
diff changeset
274
046dbe557911 write to tmp file implemented
Henry S. Thompson <ht@inf.ed.ac.uk>
parents:
diff changeset
275 whole=not (pa.warc or pa.headers or pa.body)
046dbe557911 write to tmp file implemented
Henry S. Thompson <ht@inf.ed.ac.uk>
parents:
diff changeset
276 if pa.save:
046dbe557911 write to tmp file implemented
Henry S. Thompson <ht@inf.ed.ac.uk>
parents:
diff changeset
277 _output=_output_tmpfile
046dbe557911 write to tmp file implemented
Henry S. Thompson <ht@inf.ed.ac.uk>
parents:
diff changeset
278 import tempfile
046dbe557911 write to tmp file implemented
Henry S. Thompson <ht@inf.ed.ac.uk>
parents:
diff changeset
279 elif pa.cmd:
046dbe557911 write to tmp file implemented
Henry S. Thompson <ht@inf.ed.ac.uk>
parents:
diff changeset
280 _output = _output_subproc
046dbe557911 write to tmp file implemented
Henry S. Thompson <ht@inf.ed.ac.uk>
parents:
diff changeset
281 else:
046dbe557911 write to tmp file implemented
Henry S. Thompson <ht@inf.ed.ac.uk>
parents:
diff changeset
282 _output = _output_stdout
21
cbac7dfe2f24 interpolate process0, support permanent subproc
Henry S. Thompson <ht@inf.ed.ac.uk>
parents: 18
diff changeset
283 if pa.cmd and pa.process:
cbac7dfe2f24 interpolate process0, support permanent subproc
Henry S. Thompson <ht@inf.ed.ac.uk>
parents: 18
diff changeset
284 launch(pa.cmd)
cbac7dfe2f24 interpolate process0, support permanent subproc
Henry S. Thompson <ht@inf.ed.ac.uk>
parents: 18
diff changeset
285 # three different ways to process
18
046dbe557911 write to tmp file implemented
Henry S. Thompson <ht@inf.ed.ac.uk>
parents:
diff changeset
286 if pa.index:
298
fdec28613df3 robotstxt now working?
Henry S. Thompson <ht@inf.ed.ac.uk>
parents: 238
diff changeset
287 CDX=regex.compile('length": "([0-9]*)", "offset": "([0-9]*)", "filename": "crawl-data/([^/]*)/segments/([^/]*)/(warc|crawldiagnostics|robotstxt)/(.*\.gz)"') # robotstxt works?
18
046dbe557911 write to tmp file implemented
Henry S. Thompson <ht@inf.ed.ac.uk>
parents:
diff changeset
288 for l in sys.stdin:
046dbe557911 write to tmp file implemented
Henry S. Thompson <ht@inf.ed.ac.uk>
parents:
diff changeset
289 m=CDX.search(l)
046dbe557911 write to tmp file implemented
Henry S. Thompson <ht@inf.ed.ac.uk>
parents:
diff changeset
290 if m is None:
046dbe557911 write to tmp file implemented
Henry S. Thompson <ht@inf.ed.ac.uk>
parents:
diff changeset
291 if l.find('/robotstxt/')>-1:
046dbe557911 write to tmp file implemented
Henry S. Thompson <ht@inf.ed.ac.uk>
parents:
diff changeset
292 continue
046dbe557911 write to tmp file implemented
Henry S. Thompson <ht@inf.ed.ac.uk>
parents:
diff changeset
293 print("index line problem: \"%s\""%l,file=sys.stderr,end='')
046dbe557911 write to tmp file implemented
Henry S. Thompson <ht@inf.ed.ac.uk>
parents:
diff changeset
294 exit(2)
21
cbac7dfe2f24 interpolate process0, support permanent subproc
Henry S. Thompson <ht@inf.ed.ac.uk>
parents: 18
diff changeset
295 filename=pa.fpath%(m[3:7])
cbac7dfe2f24 interpolate process0, support permanent subproc
Henry S. Thompson <ht@inf.ed.ac.uk>
parents: 18
diff changeset
296 process(pa,buf,filename,
cbac7dfe2f24 interpolate process0, support permanent subproc
Henry S. Thompson <ht@inf.ed.ac.uk>
parents: 18
diff changeset
297 int(offset:=m[2]),int(length:=m[1]),whole)
18
046dbe557911 write to tmp file implemented
Henry S. Thompson <ht@inf.ed.ac.uk>
parents:
diff changeset
298 elif pa.length is not None:
046dbe557911 write to tmp file implemented
Henry S. Thompson <ht@inf.ed.ac.uk>
parents:
diff changeset
299 print(pa.filename,file=sys.stderr)
046dbe557911 write to tmp file implemented
Henry S. Thompson <ht@inf.ed.ac.uk>
parents:
diff changeset
300 process(pa,buf,pa.fpath%tuple(pa.filename.split('/')),
046dbe557911 write to tmp file implemented
Henry S. Thompson <ht@inf.ed.ac.uk>
parents:
diff changeset
301 pa.offset,pa.length,whole)
046dbe557911 write to tmp file implemented
Henry S. Thompson <ht@inf.ed.ac.uk>
parents:
diff changeset
302 else:
046dbe557911 write to tmp file implemented
Henry S. Thompson <ht@inf.ed.ac.uk>
parents:
diff changeset
303 print("Reading length, offset, filename tab-delimited triples from stdin...",
046dbe557911 write to tmp file implemented
Henry S. Thompson <ht@inf.ed.ac.uk>
parents:
diff changeset
304 file=sys.stderr)
046dbe557911 write to tmp file implemented
Henry S. Thompson <ht@inf.ed.ac.uk>
parents:
diff changeset
305 for l in sys.stdin:
046dbe557911 write to tmp file implemented
Henry S. Thompson <ht@inf.ed.ac.uk>
parents:
diff changeset
306 try:
046dbe557911 write to tmp file implemented
Henry S. Thompson <ht@inf.ed.ac.uk>
parents:
diff changeset
307 (length,offset,filename)=l.rstrip().split('\t')
046dbe557911 write to tmp file implemented
Henry S. Thompson <ht@inf.ed.ac.uk>
parents:
diff changeset
308 length=int(length)
046dbe557911 write to tmp file implemented
Henry S. Thompson <ht@inf.ed.ac.uk>
parents:
diff changeset
309 offset=int(offset)
046dbe557911 write to tmp file implemented
Henry S. Thompson <ht@inf.ed.ac.uk>
parents:
diff changeset
310 except ValueError as e:
046dbe557911 write to tmp file implemented
Henry S. Thompson <ht@inf.ed.ac.uk>
parents:
diff changeset
311 parser.error('Invalid input line: %s\n "%s"'%(e,l))
046dbe557911 write to tmp file implemented
Henry S. Thompson <ht@inf.ed.ac.uk>
parents:
diff changeset
312 process(pa,buf,pa.fpath%tuple(filename.split('/')),
046dbe557911 write to tmp file implemented
Henry S. Thompson <ht@inf.ed.ac.uk>
parents:
diff changeset
313 offset,length,whole)
21
cbac7dfe2f24 interpolate process0, support permanent subproc
Henry S. Thompson <ht@inf.ed.ac.uk>
parents: 18
diff changeset
314 # processing done one way or another
cbac7dfe2f24 interpolate process0, support permanent subproc
Henry S. Thompson <ht@inf.ed.ac.uk>
parents: 18
diff changeset
315 if pa.cmd and pa.process:
cbac7dfe2f24 interpolate process0, support permanent subproc
Henry S. Thompson <ht@inf.ed.ac.uk>
parents: 18
diff changeset
316 windup(length,offset,filename)
cbac7dfe2f24 interpolate process0, support permanent subproc
Henry S. Thompson <ht@inf.ed.ac.uk>
parents: 18
diff changeset
317 # if pa.save and pa.process, deleting temp files is down to cmd
18
046dbe557911 write to tmp file implemented
Henry S. Thompson <ht@inf.ed.ac.uk>
parents:
diff changeset
318 if __name__ == "__main__":
046dbe557911 write to tmp file implemented
Henry S. Thompson <ht@inf.ed.ac.uk>
parents:
diff changeset
319 main()