From 24714c8784f60d3f3feaca267e781d221ef69097 Mon Sep 17 00:00:00 2001
From: "sumit.ray@baesystems.com" <sumit.ray@baesystems.com>
Date: Fri, 9 Jul 2021 22:16:37 -0400
Subject: [PATCH] Added the page node pointer to tht text neodes
---
pdf.c | 58 +++++++++++++++++++++++++++++-----------------------------
1 file changed, 29 insertions(+), 29 deletions(-)
diff --git a/pdf.c b/pdf.c
index f5436aa..efae767 100644
--- a/pdf.c
+++ b/pdf.c
@@ -122,7 +122,7 @@ typedef HCountedArray Dict;
// Catalog Tree
-typedef struct RsrcDict_T {
+typedef struct RsrcDict_S {
const HParsedToken *resources; // font references dictionary (resources == NULL) ==> inherit
const HParsedToken **fonts; // Serialized list of fonts used in this page
size_t numFonts;
@@ -130,12 +130,12 @@ typedef struct RsrcDict_T {
// size_t numSeenFonts;
const HParsedToken **seenCmaps; // memoized cmaps (should this be a bytestream?
size_t numCmapsSeen;
-} RsrcDict_S;
-struct PtNode_T;
-typedef struct PtNode_T {
+} RsrcDict_T;
+struct PtNode_S;
+typedef struct PtNode_S {
enum {PG_TREE, PG_NODE, XO_NODE} type;
const HParsedToken *parent; // Type = Page tree -- reference
- RsrcDict_S *pgRsrc; // resource structure
+ RsrcDict_T *pgRsrc; // resource structure
const HParsedToken *me; // Reference for me
size_t offset; //
union {
@@ -144,20 +144,20 @@ typedef struct PtNode_T {
const HParsedToken *textStream; // content stream -- may be a result of concatenating array of content streams
} pn;
struct {
- struct PtNode_T *kids; // page table
+ struct PtNode_S *kids; // page table
size_t count; // number of kids
size_t leaves; // number of pages in tree
} pt;
};
-} PtNode_S;
+} PtNode_T;
-typedef struct {
+typedef struct Catalog_S {
const HParsedToken *catalog; // reference
const HParsedToken *pRoot; // reference
- PtNode_S pgTree;
+ PtNode_T pgTree;
size_t pgCount;
-} Catalog_S;
+} Catalog_T;
// Forward declaration of text extraction related structures
@@ -281,7 +281,7 @@ struct Env {
struct textnode *txttail; /* parsed text objects from the file */
size_t ntextobjs;
- Catalog_S catalog; /* Catalog object and document structure */
+ Catalog_T catalog; /* Catalog object and document structure */
TextState_T tstate; /* current text state */
};
@@ -1845,8 +1845,8 @@ HParser *
ktxtstream(HAllocator *mm__, const HParsedToken *x, void *env)
{
- struct Env *aux = env;
-
+ struct Env *aux = env;
+ struct textnode *txtnd;
assert (x->token_type == TT_SEQUENCE);
int n_tobjs = x->seq->used;
@@ -1880,10 +1880,7 @@ ktxtstream(HAllocator *mm__, const HParsedToken *x, void *env)
// store the string in the environment
- // not sure whether we need to actually store the string in malloc'ed area
- // currently, we are reusing the token memory previously created
- struct textnode *txtnd = (struct textnode *) malloc(
- sizeof(struct textnode));
+ txtnd = h_alloc(mm__, sizeof(struct textnode));
txtnd->tstr = tstr;
txtnd->next = NULL;
if (aux->txthead == NULL)
@@ -3806,7 +3803,7 @@ HParsedToken *create_strmdict(HArena *arena, size_t len)
const HParsedToken *
-lookup_font(PtNode_S *page, char *name)
+lookup_font(PtNode_T *page, char *name)
{
const HParsedToken *tok = NULL;
Dict *font;
@@ -3831,7 +3828,7 @@ lookup_font(PtNode_S *page, char *name)
return (tok);
}
-void parse_fonts(HArena *arena, const HParsedToken *dictT, RsrcDict_S *pgRsrc)
+void parse_fonts(HArena *arena, const HParsedToken *dictT, RsrcDict_T *pgRsrc)
{
Dict *fontdict = H_CAST(Dict, dictT);
const HParsedToken *item = NULL;
@@ -3861,14 +3858,14 @@ void parse_fonts(HArena *arena, const HParsedToken *dictT, RsrcDict_S *pgRsrc)
}
}
-void parse_rsrcdict(HArena *arena, const HParsedToken *dictT, PtNode_S *pgNode)
+void parse_rsrcdict(HArena *arena, const HParsedToken *dictT, PtNode_T *pgNode)
{
- RsrcDict_S *rsrc = NULL;
+ RsrcDict_T *rsrc = NULL;
// Process the dictionary
if ( (dictT->token_type == TT_Dict) || (dictT->token_type == TT_Objstm) ) {
- rsrc = h_arena_malloc(arena, sizeof(RsrcDict_S));
+ rsrc = h_arena_malloc(arena, sizeof(RsrcDict_T));
rsrc->resources = dictT;
rsrc->fonts = NULL;
rsrc->numFonts = 0;
@@ -3915,7 +3912,7 @@ void parse_rsrcdict(HArena *arena, const HParsedToken *dictT, PtNode_S *pgNode)
void parse_pagenode (
struct Env *aux,
- PtNode_S *myNode,
+ PtNode_T *myNode,
const HParsedToken *myRef, // my page node reference
const Dict *myDict, // my page node specification
const HParsedToken *parent,
@@ -3933,9 +3930,12 @@ void parse_pagenode (
size_t sz = 0, nOffset = 0;
- myNode->type = PG_NODE;
- myNode->me = myRef;
- myNode->pn.dict = myDict;
+ // set some global state variables
+ aux->tstate.page = myNode;
+ myNode->type = PG_NODE;
+ myNode->me = myRef;
+ myNode->pn.dict = myDict;
+
item = dictentry(myDict, "Parent");
@@ -4068,7 +4068,7 @@ end:
void
parse_pagetree(
struct Env *aux,
- PtNode_S *myNode,
+ PtNode_T *myNode,
const HParsedToken *myRef, // my page tree node reference
const Dict *myDict, // my page tree specification
const HParsedToken *pRefT, // parent reference token
@@ -4078,7 +4078,7 @@ parse_pagetree(
const HParsedToken *item = NULL;
const HParsedToken *kids = NULL;
- PtNode_S *kid = NULL;
+ PtNode_T *kid = NULL;
const HParsedToken *kidRef = NULL; // page tree or page node reference
const HParsedToken *kidDict_t = NULL;
const Dict *kidDict = NULL;
@@ -4119,7 +4119,7 @@ parse_pagetree(
// get the kids (pgTable)
HCountedArray *pgTable = H_CAST_SEQ(kids);
size_t pgtSz = pgTable->used;
- myNode->pt.kids = (PtNode_S*)h_arena_malloc(pgTable->arena, pgtSz * sizeof(PtNode_S));
+ myNode->pt.kids = (PtNode_T*)h_arena_malloc(pgTable->arena, pgtSz * sizeof(PtNode_T));
myNode->pt.count = pgtSz;
// Process the kids
--
GitLab