make python recoll a package with 2 modules recoll and rclextract

src/lib/Makefile

@@ -13,6 +13,11 @@ librcl.a : $(DEPS) $(OBJS)
 	ar ru librcl.a $(OBJS)
 	$(RANLIB) librcl.a
 
+# Future 
+#all: librecoll.so
+#librecoll.so : $(DEPS) $(OBJS)
+#	g++ -shared -Wl,--no-undefined #          -Wl,-soname=librecoll.so.1 -o librecoll.so $(OBJS) #           -lxapian -lz -lX11 -lpthread -ldl
+
 rclaspell.o : ../aspell/rclaspell.cpp $(depth)/mk/localdefs
 	$(CXX) $(ALL_CXXFLAGS) -c ../aspell/rclaspell.cpp
 beaglequeuecache.o : ../common/beaglequeuecache.cpp $(depth)/mk/localdefs

src/lib/mkMake

@@ -126,6 +126,13 @@ librcl.a : \$(DEPS) \$(OBJS)
 	ar ru librcl.a \$(OBJS)
 	\$(RANLIB) librcl.a
 
+# Future 
+#all: librecoll.so
+#librecoll.so : \$(DEPS) \$(OBJS)
+#	g++ -shared -Wl,--no-undefined \
+#          -Wl,-soname=librecoll.so.1 -o librecoll.so \$(OBJS) \
+#           -lxapian -lz -lX11 -lpthread -ldl
+
 EOF
 
 for c in $SRC_CPP;do

src/python/recoll/pyrclextract.cpp

@@ -178,9 +178,15 @@ Extractor_idoctofile(rclx_ExtractorObject* self, PyObject *args,
 	outfile.assign(soutfile); 
     
     if (self->xtr == 0) {
-        PyErr_SetString(PyExc_AttributeError, "extract: null object");
+        PyErr_SetString(PyExc_AttributeError, "idoctofile: null object");
 	return 0;
     }
+    if (ipath.empty()) {
+        PyErr_SetString(PyExc_ValueError, "idoctofile: null ipath");
+	return 0;
+    }
+	
+    self->xtr->setTargetMType(mimetype);
     TempFile temp;
     bool status = self->xtr->interntofile(temp, outfile, ipath, mimetype);
     if (!status) {
@@ -285,5 +291,5 @@ initrclextract(void)
     Py_INCREF(&rclx_ExtractorType);
     PyModule_AddObject(m, "Extractor", (PyObject *)&rclx_ExtractorType);
 
-    recoll_DocType = (PyObject*)PyCapsule_Import("recoll.doctypeptr", 0);
+    recoll_DocType = (PyObject*)PyCapsule_Import(PYRECOLL_PACKAGE "recoll.doctypeptr", 0);
 }

src/python/recoll/pyrecoll.cpp

@@ -1703,6 +1703,6 @@ initrecoll(void)
 
     // Export a few pointers for the benefit of other recoll python modules
     PyObject* doctypecapsule = 
-	PyCapsule_New(&recoll_DocType, "recoll.doctypeptr", 0);
+	PyCapsule_New(&recoll_DocType, PYRECOLL_PACKAGE "recoll.doctypeptr", 0);
     PyModule_AddObject(m, "doctypeptr", doctypecapsule);
 }

src/python/recoll/pyrecoll.h

@@ -28,4 +28,6 @@ typedef struct {
     RclConfig *rclconfig; 
 } recoll_DocObject;
 
+#define PYRECOLL_PACKAGE "recoll."
+
 #endif // _PYRECOLL_H_INCLUDED_

src/python/recoll/setup.py.in

@@ -50,8 +50,7 @@ module1 = Extension('recoll',
                                     ],
                     libraries = libs,
                     library_dirs = libdirs,
-                    sources = ['pyrecoll.cpp',
+                    sources = ['pyrecoll.cpp'])
-                               ])
 
 module2 = Extension('rclextract',
                     define_macros = [('MAJOR_VERSION', '1'),
@@ -67,14 +66,17 @@ module2 = Extension('rclextract',
                                     ],
                     libraries = libs,
                     library_dirs = libdirs,
-                    sources = ['pyrclextract.cpp',
+                    sources = ['pyrclextract.cpp'])
-                               ])
 
 setup (name = 'Recoll',
        version = '1.0',
        description = 'Query/Augment a Recoll full text index',
        author = 'J.F. Dockes',
        author_email = 'jfd@recoll.org',
+       url = 'http://www.recoll.org',
+       license = 'GPL',
        long_description = '''
 ''',
+    packages = ['recoll'],
+    ext_package = 'recoll',
        ext_modules = [module1, module2])

src/python/samples/recollq.py

@@ -6,8 +6,14 @@ This could actually be useful for something after some customization
 
 import sys
 from getopt import getopt
-import recoll
+
-import rclextract
+try:
+    from recoll import recoll
+    from recoll import rclextract
+    hasextract = True
+except:
+    import recoll
+    hasextract = False
     
 allmeta = ("title", "keywords", "abstract", "url", "mimetype", "mtime",
            "ipath", "fbytes", "dbytes", "relevancyrating")
@@ -27,9 +33,15 @@ class ptrmeths:
     
 def extract(doc):
     extractor = rclextract.Extractor(doc)
-    newdoc = extractor.extract(doc.ipath)
+    newdoc = extractor.textextract(doc.ipath)
     return newdoc
 
+def extractofile(doc, outfilename=""):
+    extractor = rclextract.Extractor(doc)
+    outfilename = extractor.idoctofile(doc.ipath, doc.mimetype, \
+                                       ofilename=outfilename)
+    return outfilename
+
 def doquery(db, q):
     # Get query object
     query = db.query()
@@ -48,9 +60,11 @@ def doquery(db, q):
     while query.next >= 0 and query.next < nres: 
         doc = query.fetchone()
         print query.next, ":",
-#        for k,v in doc.items().items():
+        #for k,v in doc.items().items():
-#            print "KEY:", k.encode('utf-8'), "VALUE", v.encode('utf-8')
+        #print "KEY:", k.encode('utf-8'), "VALUE", v.encode('utf-8')
-#        continue
+        #continue
+        #outfile = extractofile(doc)
+        #print "outfile:", outfile, "url", doc.url.encode("utf-8")
         for k in ("title", "mtime", "author"):
             value = getattr(doc, k)
 #            value = doc.get(k)