[python] Fork of the suffix-array extractor with surface improvements

Available as the cdec.sa module, with commande-line helpers:
python -m cdec.sa.compile -f ... -e ... -a ... -o sa-out/ -c extract.ini
python -m cdec.sa.extract -c extract.ini -g grammars-out/ < input.txt > input.sgml

+ renamed cdec.scfg -> cdec.sa
+ Python README

1 files changed, 23 insertions, 0 deletions

diff --git a/python/src/sa/str_map.pxi b/python/src/sa/str_map.pxi
new file mode 100644
index 00000000..3a8439a4
--- /dev/null
+++ b/python/src/sa/str_map.pxi
@@ -0,0 +1,23 @@
+cdef extern from "strmap.h":
+    ctypedef struct StrMap
+    StrMap* stringmap_new()
+    void stringmap_delete(StrMap *vocab)
+    int stringmap_index(StrMap *vocab, char *s)
+    char* stringmap_word(StrMap *vocab, int i)
+
+cdef class StringMap:
+    cdef StrMap *vocab
+    cdef char *word(self, int i)
+    cdef int index(self, char *s)
+
+    def __cinit__(self):
+        self.vocab = stringmap_new()
+
+    def __dealloc__(self):
+        stringmap_delete(self.vocab)
+        
+    cdef char *word(self, int i):
+        return stringmap_word(self.vocab, i)
+
+    cdef int index(self, char *s):
+        return stringmap_index(self.vocab, s)