info file interner
This commit is contained in:
Jean-Francois Dockes
parent
4607a1fcb8
commit
e4b1b5c9d8
227
src/filters/rclinfo
Executable file
227
src/filters/rclinfo
Executable file
@ -0,0 +1,227 @@
|
|||||||
|
#!/usr/bin/env python
|
||||||
|
|
||||||
|
# Read a file in GNU info format and output its nodes as subdocs,
|
||||||
|
# interfacing with recoll execm
|
||||||
|
|
||||||
|
|
||||||
|
import rclexecm
|
||||||
|
import sys
|
||||||
|
import os.path
|
||||||
|
import subprocess
|
||||||
|
|
||||||
|
# Prototype for the html document we're returning. Info files are
|
||||||
|
# normally ascii. Set no charset, and let it be provided by the
|
||||||
|
# environment if necessary
|
||||||
|
htmltemplate = '''
|
||||||
|
<html>
|
||||||
|
<head>
|
||||||
|
<title>%s</title>
|
||||||
|
</head>
|
||||||
|
<body><pre>
|
||||||
|
%s
|
||||||
|
</pre></body>
|
||||||
|
</html>
|
||||||
|
'''
|
||||||
|
|
||||||
|
# RclExecm interface
|
||||||
|
class InfoExtractor:
|
||||||
|
def __init__(self, em):
|
||||||
|
self.file = ""
|
||||||
|
self.contents = []
|
||||||
|
self.em = em
|
||||||
|
self.em.setmimetype("text/html")
|
||||||
|
|
||||||
|
def extractone(self, index):
|
||||||
|
if index >= len(self.contents):
|
||||||
|
return(False, "", "", True)
|
||||||
|
|
||||||
|
nodename, docdata = self.contents[index]
|
||||||
|
nodename = self.em.htmlescape(nodename)
|
||||||
|
docdata = self.em.htmlescape(docdata)
|
||||||
|
|
||||||
|
docdata = htmltemplate % (nodename, docdata)
|
||||||
|
|
||||||
|
iseof = rclexecm.RclExecM.noteof
|
||||||
|
if self.currentindex >= len(self.contents) -1:
|
||||||
|
iseof = rclexecm.RclExecM.eofnext
|
||||||
|
return (True, docdata, nodename, iseof)
|
||||||
|
|
||||||
|
###### File type handler api, used by rclexecm ---------->
|
||||||
|
def openfile(self, params):
|
||||||
|
self.file = params["filename:"]
|
||||||
|
|
||||||
|
if not os.path.isfile(self.file):
|
||||||
|
self.em.rclog("Openfile: %s is not a file" % self.file)
|
||||||
|
return False
|
||||||
|
|
||||||
|
cmd = "info --subnodes -o - -f " + self.file
|
||||||
|
nullstream = open("/dev/null", 'w')
|
||||||
|
try:
|
||||||
|
infostream = subprocess.Popen(cmd, shell=True, bufsize=1,
|
||||||
|
stderr=nullstream,
|
||||||
|
stdout=subprocess.PIPE).stdout
|
||||||
|
except Exception, e:
|
||||||
|
# Consider this as permanently fatal.
|
||||||
|
self.em.rclog("Openfile: exec info: %s" % str(e))
|
||||||
|
print "RECFILTERROR HELPERNOTFOUND info"
|
||||||
|
sys.exit(1);
|
||||||
|
|
||||||
|
|
||||||
|
self.currentindex = 0
|
||||||
|
|
||||||
|
self.contents = InfoSimpleSplitter().splitinfo(self.file, infostream)
|
||||||
|
|
||||||
|
#self.em.rclog("openfile: Entry count: %d"%(len(self.contents)))
|
||||||
|
return True
|
||||||
|
|
||||||
|
# Extract specific node
|
||||||
|
def getipath(self, params):
|
||||||
|
try:
|
||||||
|
nodename = params["ipath:"]
|
||||||
|
except:
|
||||||
|
return False
|
||||||
|
|
||||||
|
# We could build a dictionary in the split function to avoid this.
|
||||||
|
# But it's used for preview, and the perf issue is minimal
|
||||||
|
for i in range(len(self.contents)):
|
||||||
|
if self.contents[i][0] == nodename:
|
||||||
|
return self.extractone(i)
|
||||||
|
return (False, "", "", True)
|
||||||
|
|
||||||
|
# Extract next in list
|
||||||
|
def getnext(self, params):
|
||||||
|
if self.currentindex >= len(self.contents):
|
||||||
|
self.em.rclog("getnext: EOF hit")
|
||||||
|
return (False, "", "", rclexecm.RclExecM.eofnow)
|
||||||
|
else:
|
||||||
|
ret= self.extractone(self.currentindex)
|
||||||
|
self.currentindex += 1
|
||||||
|
return ret
|
||||||
|
|
||||||
|
# Info file splitter
|
||||||
|
class InfoSimpleSplitter:
|
||||||
|
|
||||||
|
def splitinfo(self, filename, fin):
|
||||||
|
gotblankline = 1
|
||||||
|
index = 0
|
||||||
|
listout = []
|
||||||
|
node_dict = {}
|
||||||
|
node = ""
|
||||||
|
|
||||||
|
for line in fin:
|
||||||
|
if gotblankline and line.startswith("File: "):
|
||||||
|
if index != 0:
|
||||||
|
listout.append((nodename, node))
|
||||||
|
line = line.rstrip("\n\r")
|
||||||
|
pairs = line.split(",")
|
||||||
|
up = "Top"
|
||||||
|
nodename = str(index)
|
||||||
|
infofile = os.path.basename(filename)
|
||||||
|
for pair in pairs:
|
||||||
|
name, value = pair.split(':')
|
||||||
|
name = name.strip(" ")
|
||||||
|
value = value.replace(":", " ");
|
||||||
|
value = value.replace("|", " ").strip(" ")
|
||||||
|
if name == "Node":
|
||||||
|
nodename = value
|
||||||
|
if name == "Up":
|
||||||
|
up = value
|
||||||
|
if name == "File":
|
||||||
|
infofile = value
|
||||||
|
|
||||||
|
if node_dict.has_key(nodename):
|
||||||
|
print >> sys.stderr, "Info file", filename, \
|
||||||
|
"Dup node: ", nodename
|
||||||
|
node_dict[nodename] = up
|
||||||
|
print "NODE ", nodename, "UP ", up
|
||||||
|
node = ""
|
||||||
|
index += 1
|
||||||
|
|
||||||
|
if line.rstrip("\n\r") == '':
|
||||||
|
gotblankline = 1
|
||||||
|
else:
|
||||||
|
gotblankline = 0
|
||||||
|
|
||||||
|
node += line
|
||||||
|
|
||||||
|
# File done, add last dangling node
|
||||||
|
if node != "":
|
||||||
|
listout.append((nodename, node))
|
||||||
|
|
||||||
|
# Compute node paths (concatenate "Up" values), to be used
|
||||||
|
# as page titles and ipaths. It's unfortunate that this will crash if
|
||||||
|
# the info file tree is bad
|
||||||
|
listout1 = []
|
||||||
|
for nodename, node in listout:
|
||||||
|
ipath = ""
|
||||||
|
loop = 0
|
||||||
|
error = 0
|
||||||
|
while nodename != "Top":
|
||||||
|
ipath = nodename + " / " + ipath
|
||||||
|
if node_dict.has_key(nodename):
|
||||||
|
nodename = node_dict[nodename]
|
||||||
|
else:
|
||||||
|
print >> sys.stderr, \
|
||||||
|
"Infofile: node's Up does not exist: file %s, path %s, up [%s]" % \
|
||||||
|
(infofile, ipath, nodename)
|
||||||
|
error = 1
|
||||||
|
break
|
||||||
|
loop += 1
|
||||||
|
if loop > 50:
|
||||||
|
print >> sys.stderr, "Infofile: bad tree (looping)", \
|
||||||
|
infofile
|
||||||
|
error = 1
|
||||||
|
break
|
||||||
|
|
||||||
|
if error:
|
||||||
|
continue
|
||||||
|
|
||||||
|
if ipath == "":
|
||||||
|
ipath = infofile
|
||||||
|
else:
|
||||||
|
ipath = infofile + " / " + ipath
|
||||||
|
ipath = ipath.rstrip(" / ")
|
||||||
|
listout1.append((ipath, node))
|
||||||
|
|
||||||
|
return listout1
|
||||||
|
|
||||||
|
|
||||||
|
##### Main program: either talk to the parent or execute test loop
|
||||||
|
|
||||||
|
e = rclexecm.RclExecM()
|
||||||
|
info = InfoExtractor(e)
|
||||||
|
|
||||||
|
if len(sys.argv) == 1:
|
||||||
|
e.mainloop(info)
|
||||||
|
else:
|
||||||
|
# Got a file name parameter: TESTING without an execm parent
|
||||||
|
# Loop on all entries or get specific ipath
|
||||||
|
if not info.openfile({'filename:':sys.argv[1]}):
|
||||||
|
print "Open error"
|
||||||
|
sys.exit(1)
|
||||||
|
ipath = ""
|
||||||
|
if len(sys.argv) == 3:
|
||||||
|
ipath = sys.argv[2]
|
||||||
|
|
||||||
|
if ipath != "":
|
||||||
|
ok, data, ipath, eof = info.getipath({'ipath:':ipath})
|
||||||
|
if ok:
|
||||||
|
print "=========== ENTRY for IPATH %s =============" % (ipath)
|
||||||
|
print data
|
||||||
|
print
|
||||||
|
else:
|
||||||
|
print "Got error, eof %d"%eof
|
||||||
|
sys.exit(0)
|
||||||
|
|
||||||
|
ecnt = 0
|
||||||
|
while 1:
|
||||||
|
ok, data, ipath, eof = info.getnext("")
|
||||||
|
if ok:
|
||||||
|
ecnt = ecnt + 1
|
||||||
|
print "=========== ENTRY %d IPATH %s =============" % (ecnt,ipath)
|
||||||
|
# print data
|
||||||
|
print
|
||||||
|
else:
|
||||||
|
print "Got error, eof %d"%eof
|
||||||
|
break
|
||||||
|
|
||||||
Loading…
x
Reference in New Issue
Block a user