Go through all references to the signature dictionary

include/acroform.h

+#ifndef PDF_SIGIL_ACROFORM_H
+#define PDF_SIGIL_ACROFORM_H
+
+#include "types.h"
+
+
+sigil_err_t process_acroform(sigil_t *sgl);
+
+int sigil_auxiliary_self_test(int verbosity);
+
+#endif //PDF_SIGIL_ACROFORM_H

include/auxiliary.h

@@ -20,6 +20,9 @@ sigil_err_t pdf_peek_char(sigil_t *sgl, char *result);
 
 sigil_err_t pdf_move_pos_rel(sigil_t *sgl, ssize_t shift_bytes);
 sigil_err_t pdf_move_pos_abs(sigil_t *sgl, size_t position);
+sigil_err_t pdf_goto_obj(sigil_t *sgl, reference_t *ref);
+
+sigil_err_t get_curr_position(sigil_t *sgl, size_t *result);
 
 sigil_err_t skip_leading_whitespaces(sigil_t *sgl);
 sigil_err_t skip_array(sigil_t *sgl);
@@ -30,6 +33,9 @@ sigil_err_t parse_number(sigil_t *sgl, size_t *number);
 sigil_err_t parse_word(sigil_t *sgl, const char *word);
 sigil_err_t parse_indirect_reference(sigil_t *sgl, reference_t *ref);
 sigil_err_t parse_dict_key(sigil_t *sgl, dict_key_t *dict_key);
+sigil_err_t parse_ref_array(sigil_t *sgl, ref_array_t *ref_array);
+
+sigil_err_t reference_to_offset(sigil_t *sgl, const reference_t *ref, size_t *result);
 
 const char *sigil_err_string(sigil_err_t err);
 

include/catalog_dict.h

+#ifndef PDF_SIGIL_CATALOG_DICT_H
+#define PDF_SIGIL_CATALOG_DICT_H
+
+#include "types.h"
+
+
+sigil_err_t process_catalog_dictionary(sigil_t *sgl);
+
+int sigil_header_self_test(int verbosity);
+
+#endif /* PDF_SIGIL_CATALOG_DICT_H */

include/config.h

@@ -11,6 +11,9 @@
 // capacity to choose for the first xref allocation
 #define XREF_PREALLOCATION          10
 
+// capacity to choose for the first allocation in array of fields
+#define REF_ARRAY_PREALLOCATION        10
+
 // threshold in bytes for loading whole file into buffer
 #define THRESHOLD_FILE_BUFFERING    10485760
 

include/constants.h

@@ -3,27 +3,39 @@
 
 
 #ifdef __unix__
-    #define COLOR_RED          "\x1b[31m"
-    #define COLOR_GREEN        "\x1b[32m"
-    #define COLOR_RESET        "\x1b[0m"
+    #define COLOR_RED         "\x1b[31m"
+    #define COLOR_GREEN       "\x1b[32m"
+    #define COLOR_RESET       "\x1b[0m"
 #else
-    #define COLOR_RED          ""
-    #define COLOR_GREEN        ""
-    #define COLOR_RESET        ""
+    #define COLOR_RED         ""
+    #define COLOR_GREEN       ""
+    #define COLOR_RESET       ""
 #endif
 
 
-#define XREF_TYPE_UNSET    0
-#define XREF_TYPE_TABLE    1
-#define XREF_TYPE_STREAM   2
+#define XREF_TYPE_UNSET       0
+#define XREF_TYPE_TABLE       1
+#define XREF_TYPE_STREAM      2
 
-#define DICT_KEY_UNKNOWN   0
-#define DICT_KEY_Size      1
-#define DICT_KEY_Prev      2
-#define DICT_KEY_Root      3
+#define DICT_KEY_UNKNOWN      0
+#define DICT_KEY_Size         1
+#define DICT_KEY_Prev         2
+#define DICT_KEY_Root         3
+#define DICT_KEY_AcroForm     4
+#define DICT_KEY_Fields       5
+#define DICT_KEY_SigFlags     6
+#define DICT_KEY_FT           7
+#define DICT_KEY_V            8
+#define DICT_KEY_SubFilter    9
+#define DICT_KEY_Cert         10
+#define DICT_KEY_Contents     11
+#define DICT_KEY_ByteRange    12
 
-#define DEALLOCATE_FILE    0x01
-#define DEALLOCATE_BUFFER  0x02
+#define SUBFILTER_UNKNOWN               0
+#define SUBFILTER_adbe_x509_rsa_sha1    1
+
+#define DEALLOCATE_FILE       0x01
+#define DEALLOCATE_BUFFER     0x02
 
 #define ERR_NO                0
 #define ERR_ALLOCATION        1
@@ -33,5 +45,6 @@
 #define ERR_NOT_IMPLEMENTED   5
 #define ERR_NO_DATA           6
 #define ERR_END_OF_DICT       7
+#define ERR_NO_SIGNATURE      8
 
 #endif /* PDF_SIGIL_CONSTANTS_H */

include/sig_dict.h

+#ifndef PDF_SIGIL_SIG_DICT_H
+#define PDF_SIGIL_SIG_DICT_H
+
+#include "types.h"
+
+
+sigil_err_t process_sig_dict(sigil_t *sgl);
+
+int sigil_sigil_self_test(int verbosity);
+
+#endif /* PDF_SIGIL_SIG_DICT_H */

include/sig_field.h

+#ifndef PDF_SIGIL_SIG_FIELD_H
+#define PDF_SIGIL_SIG_FIELD_H
+
+#include "types.h"
+
+
+sigil_err_t find_sig_field(sigil_t *sgl);
+
+sigil_err_t process_sig_field(sigil_t *sgl);
+
+int sigil_sigil_self_test(int verbosity);
+
+#endif /* PDF_SIGIL_SIG_FIELD_H */

include/types.h

@@ -12,7 +12,7 @@
 
 typedef uint32_t sigil_err_t;
 
-typedef uint32_t keyword_t;
+typedef uint32_t subfilter_t;
 
 typedef uint32_t dict_key_t;
 
@@ -21,6 +21,17 @@ typedef struct {
     size_t generation_num;
 } reference_t;
 
+typedef struct range_t {
+    size_t start;
+    size_t length;
+    struct range_t *next;
+} range_t;
+
+typedef struct {
+    reference_t **entry;
+    size_t capacity;
+} ref_array_t;
+
 typedef struct xref_entry_t {
     size_t byte_offset;
     size_t generation_num;
@@ -49,8 +60,17 @@ typedef struct {
     short       xref_type;
     xref_t     *xref;
     reference_t ref_catalog_dict;
+    reference_t ref_acroform;
+    size_t      offset_acroform;
+    reference_t ref_sig_field;
+    reference_t ref_sig_dict;
+    size_t      offset_sig_dict;
+    ref_array_t fields;
     size_t      pdf_start_offset;  /* offset of %PDF-x.y      */
     size_t      startxref;
+    size_t      sig_flags;
+    subfilter_t subfilter;
+    range_t    *byte_range;
 } sigil_t;
 
 #endif /* PDF_SIGIL_TYPES_H */

lib/acroform.c

+#include <types.h>
+#include "acroform.h"
+#include "auxiliary.h"
+#include "constants.h"
+#include "types.h"
+
+
+sigil_err_t process_acroform(sigil_t *sgl)
+{
+    sigil_err_t err;
+    dict_key_t dict_key;
+
+    if (sgl == NULL)
+        return ERR_PARAMETER;
+
+    if (sgl->offset_acroform <= 0 && sgl->ref_acroform.object_num > 0) {
+        err = pdf_goto_obj(sgl, &(sgl->ref_acroform));
+        if (err != ERR_NO)
+            return err;
+    } else {
+        err = pdf_move_pos_abs(sgl, sgl->offset_acroform);
+        if (err != ERR_NO)
+            return err;
+    }
+
+    err = parse_word(sgl, "<<");
+    if (err != ERR_NO)
+        return err;
+
+    while ((err = parse_dict_key(sgl, &dict_key)) == ERR_NO) {
+        switch (dict_key) {
+            case DICT_KEY_Fields:
+                err = parse_ref_array(sgl, &(sgl->fields));
+                if (err != ERR_NO)
+                    return err;
+                break;
+            case DICT_KEY_SigFlags:
+                err = parse_number(sgl, &(sgl->sig_flags));
+                if (err != ERR_NO)
+                    return err;
+                break;
+            case DICT_KEY_UNKNOWN:
+                err = skip_dict_unknown_value(sgl);
+                if (err != ERR_NO)
+                    return err;
+                break;
+            default:
+                return ERR_PDF_CONTENT;
+        }
+    }
+
+    if (err == ERR_END_OF_DICT)
+        return ERR_NO;
+
+    return err;
+}
\ No newline at end of file

lib/auxiliary.c

 #include <stdio.h>
+#include <stdlib.h>
 #include <string.h>
 #include "auxiliary.h"
+#include "config.h"
 #include "constants.h"
 #include "sigil.h"
+#include "types.h"
 
-#define DICT_KEY_MAX   10
+#define DICT_KEY_MAX   20
 
 
 void sigil_zeroize(void *a, size_t bytes)
@@ -191,6 +194,70 @@ sigil_err_t pdf_move_pos_abs(sigil_t *sgl, size_t position)
     return ERR_NO_DATA;
 }
 
+sigil_err_t pdf_goto_obj(sigil_t *sgl, reference_t *ref)
+{
+    sigil_err_t err;
+    size_t offset,
+           tmp;
+
+    if (sgl == NULL || ref == NULL)
+        return ERR_PARAMETER;
+
+    if (ref->object_num <= 0 &&
+        ref->generation_num <= 0)
+    {
+        return ERR_NO_DATA;
+    }
+
+    err = reference_to_offset(sgl, ref, &offset);
+    if (err != ERR_NO)
+        return err;
+
+    err = pdf_move_pos_abs(sgl, offset);
+    if (err != ERR_NO)
+        return err;
+
+    err = parse_number(sgl, &tmp);
+    if (err != ERR_NO)
+        return err;
+    if (tmp != ref->object_num)
+        return ERR_PDF_CONTENT;
+
+    err = parse_number(sgl, &tmp);
+    if (err != ERR_NO)
+        return err;
+    if (tmp != ref->generation_num)
+        return ERR_PDF_CONTENT;
+
+    err = parse_word(sgl, "obj");
+    if (err != ERR_NO)
+        return err;
+
+    return ERR_NO;
+}
+
+sigil_err_t get_curr_position(sigil_t *sgl, size_t *result)
+{
+    if (sgl == NULL || result == NULL)
+        return ERR_PARAMETER;
+
+    if (sgl->pdf_data.buffer != NULL) {
+        *result = sgl->pdf_data.buf_pos;
+
+        return ERR_NO;
+    }
+
+    if (sgl->pdf_data.file != NULL) {
+        *result = (size_t)ftell(sgl->pdf_data.file);
+        if (*result < 0)
+            return ERR_IO;
+
+        return ERR_NO;
+    }
+
+    return ERR_NO_DATA;
+}
+
 sigil_err_t skip_leading_whitespaces(sigil_t *sgl)
 {
     sigil_err_t err;
@@ -272,26 +339,43 @@ sigil_err_t skip_dict_unknown_value(sigil_t *sgl)
 {
     sigil_err_t err;
     char c;
+    int first_non_whitespace = 1;
 
     while ((err = pdf_peek_char(sgl, &c)) == ERR_NO) {
-        switch (c) {
-            case '/':
-                return ERR_NO;
-            case '[':
-                if ((err = pdf_move_pos_rel(sgl, 1)) != ERR_NO)
-                    return err;
-                if ((err = skip_array(sgl)) != ERR_NO)
-                    return err;
-                return ERR_NO;
-            case '<':
-                if ((err = pdf_move_pos_rel(sgl, 1)) != ERR_NO)
-                    return err;
-                if ((err = pdf_peek_char(sgl, &c)) != ERR_NO)
-                    return err;
-                if (c != '<')
+        if (first_non_whitespace && !is_whitespace(c)) {
+            first_non_whitespace = 0;
+
+            switch (c) {
+                case '/':
+                    if ((err = pdf_move_pos_rel(sgl, 1)) != ERR_NO)
+                        return err;
                     break;
-                if ((err = skip_dictionary(sgl)) != ERR_NO)
-                    return err;
+                case '[':
+                    if ((err = pdf_move_pos_rel(sgl, 1)) != ERR_NO)
+                        return err;
+                    if ((err = skip_array(sgl)) != ERR_NO)
+                        return err;
+                    return ERR_NO;
+                case '<':
+                    if ((err = pdf_move_pos_rel(sgl, 1)) != ERR_NO)
+                        return err;
+                    if ((err = pdf_peek_char(sgl, &c)) != ERR_NO)
+                        return err;
+                    if (c != '<')
+                        break;
+                    if ((err = skip_dictionary(sgl)) != ERR_NO)
+                        return err;
+                    return ERR_NO;
+                default:
+                    break;
+            }
+
+            continue;
+        }
+
+        switch (c) {
+            case '/': // key of next entry
+            case '>': // end of dictionary
                 return ERR_NO;
             case EOF:
                 return ERR_PDF_CONTENT;
@@ -420,12 +504,35 @@ sigil_err_t parse_dict_key(sigil_t *sgl, dict_key_t *dict_key)
             return err;
     }
 
+    tmp[count] = '\0';
+
+    if (err != ERR_NO)
+        return err;
+
     if (strncmp(tmp, "Size", 4) == 0) {
         *dict_key = DICT_KEY_Size;
     } else if (strncmp(tmp, "Prev", 4) == 0) {
         *dict_key = DICT_KEY_Prev;
     } else if (strncmp(tmp, "Root", 4) == 0) {
         *dict_key = DICT_KEY_Root;
+    } else if (strncmp(tmp, "AcroForm", 8) == 0) {
+        *dict_key = DICT_KEY_AcroForm;
+    } else if (strncmp(tmp, "Fields", 6) == 0) {
+        *dict_key = DICT_KEY_Fields;
+    } else if (strncmp(tmp, "SigFlags", 8) == 0) {
+        *dict_key = DICT_KEY_SigFlags;
+    } else if (strncmp(tmp, "FT", 2) == 0) {
+        *dict_key = DICT_KEY_FT;
+    } else if (strncmp(tmp, "V", 1) == 0) {
+        *dict_key = DICT_KEY_V;
+    } else if (strncmp(tmp, "SubFilter", 9) == 0) {
+        *dict_key = DICT_KEY_SubFilter;
+    } else if (strncmp(tmp, "Cert", 4) == 0) {
+        *dict_key = DICT_KEY_Cert;
+    } else if (strncmp(tmp, "Contents", 8) == 0) {
+        *dict_key = DICT_KEY_Contents;
+    } else if (strncmp(tmp, "ByteRange", 9) == 0) {
+        *dict_key = DICT_KEY_ByteRange;
     } else {
         *dict_key = DICT_KEY_UNKNOWN;
     }
@@ -433,6 +540,88 @@ sigil_err_t parse_dict_key(sigil_t *sgl, dict_key_t *dict_key)
     return ERR_NO;
 }
 
+// parsing array of indirect references into ref_array
+sigil_err_t parse_ref_array(sigil_t *sgl, ref_array_t *ref_array)
+{
+    sigil_err_t err;
+    size_t position;
+    reference_t reference;
+
+    if (sgl == NULL || ref_array == NULL)
+        return ERR_PARAMETER;
+
+    if ((err = parse_word(sgl, "[")) != ERR_NO)
+        return err;
+
+    if (parse_word(sgl, "]") == ERR_NO) // empty array
+        return ERR_NO;
+
+    if (ref_array->capacity <= 0) {
+        ref_array->entry = malloc(sizeof(*ref_array->entry) * REF_ARRAY_PREALLOCATION);
+        if (ref_array->entry == NULL)
+            return ERR_ALLOCATION;
+        sigil_zeroize(ref_array->entry, sizeof(*ref_array->entry) * REF_ARRAY_PREALLOCATION);
+        ref_array->capacity = REF_ARRAY_PREALLOCATION;
+    }
+
+    position = 0;
+
+    while ((err = parse_indirect_reference(sgl,&reference)) == ERR_NO) {
+        if (position >= ref_array->capacity) {
+            ref_array->entry = realloc(ref_array->entry,
+                sizeof(*ref_array->entry) * ref_array->capacity * 2);
+
+            if (ref_array->entry == NULL)
+                return ERR_ALLOCATION;
+
+            sigil_zeroize(ref_array->entry + ref_array->capacity,
+                sizeof(*ref_array->entry) * ref_array->capacity);
+
+            ref_array->capacity *= 2;
+        }
+
+        if (ref_array->entry[position] == NULL) {
+            ref_array->entry[position] = malloc(sizeof(reference_t));
+            if (ref_array->entry[position] == NULL)
+                return ERR_ALLOCATION;
+        }
+
+        ref_array->entry[position]->object_num = reference.object_num;
+        ref_array->entry[position]->generation_num = reference.generation_num;
+
+        if (parse_word(sgl, "]") == ERR_NO)
+            return ERR_NO;
+
+        position++;
+    }
+
+    return err;
+}
+
+sigil_err_t reference_to_offset(sigil_t *sgl, const reference_t *ref, size_t *result)
+{
+    xref_entry_t *xref_entry;
+
+    if (sgl == NULL || ref == NULL || sgl->xref == NULL)
+        return ERR_PARAMETER;
+
+    if (sgl->xref->capacity <= ref->object_num)
+        return ERR_NO_DATA;
+
+    xref_entry = sgl->xref->entry[ref->object_num];
+
+    while (xref_entry != NULL) {
+        if (xref_entry->generation_num == ref->generation_num) {
+            *result = xref_entry->byte_offset;
+            return ERR_NO;
+        }
+
+        xref_entry = xref_entry->next;
+    }
+
+    return ERR_NO_DATA;
+}
+
 const char *sigil_err_string(sigil_err_t err)
 {
     switch (err) {

lib/catalog_dict.c

+#include "acroform.h"
+#include "auxiliary.h"
+#include "catalog_dict.h"
+#include "constants.h"
+#include "types.h"
+
+sigil_err_t process_catalog_dictionary(sigil_t *sgl)
+{
+    sigil_err_t err;
+    size_t offset;
+    dict_key_t dict_key;
+    char c;
+
+    if (sgl == NULL)
+        return ERR_PARAMETER;
+
+    if (sgl->ref_catalog_dict.object_num <= 0 &&
+        sgl->ref_catalog_dict.generation_num <= 0)
+    {
+        return ERR_NO_DATA;
+    }
+
+    err = pdf_goto_obj(sgl, &(sgl->ref_catalog_dict));
+    if (err != ERR_NO)
+        return err;
+
+    err = parse_word(sgl, "<<");
+    if (err != ERR_NO)
+        return err;
+
+    while ((err = parse_dict_key(sgl, &dict_key)) == ERR_NO) {
+        switch (dict_key) {
+            case DICT_KEY_AcroForm:
+                if ((err = skip_leading_whitespaces(sgl)) != ERR_NO)
+                    return err;
+
+                if ((err = pdf_peek_char(sgl, &c)) != ERR_NO)
+                    return err;
+
+                if (c == '<') {
+                    if ((err = get_curr_position(sgl, &offset)) != ERR_NO)
+                        return err;
+
+                    sgl->offset_acroform = offset;
+
+                    if ((err = parse_word(sgl, "<<")) != ERR_NO)
+                        return err;
+
+                    err = skip_dictionary(sgl);
+                    if (err != ERR_NO)
+                        return err;
+                } else {
+                    err = parse_indirect_reference(sgl, &(sgl->ref_acroform));
+                    if (err != ERR_NO)
+                        return err;
+                }
+                break;
+            case DICT_KEY_UNKNOWN:
+                err = skip_dict_unknown_value(sgl);
+                if (err != ERR_NO)
+                    return err;
+                break;
+            default:
+                return ERR_PDF_CONTENT;
+        }
+    }
+
+    if (err == ERR_END_OF_DICT)
+        return ERR_NO;
+
+    return err;
+}

lib/config.c

@@ -29,6 +29,14 @@ int sigil_config_self_test(int verbosity)
 
     print_test_result(1, verbosity);
 
+    // TEST: REF_ARRAY_PREALLOCATION
+    print_test_item("FIELDS_PREALLOCATION", verbosity);
+
+    if (REF_ARRAY_PREALLOCATION < 1)
+        goto failed;
+
+    print_test_result(1, verbosity);
+
     // TEST: THRESHOLD_FILE_BUFFERING
     print_test_item("THRESHOLD_FILE_BUFFERING", verbosity);
 

lib/sig_dict.c

+#include <stdlib.h>
+#include <string.h>
+#include <types.h>
+#include "auxiliary.h"
+#include "constants.h"
+#include "sig_dict.h"
+
+
+#define SUBFILTER_MAX    30
+
+static sigil_err_t parse_subfilter(sigil_t *sgl)
+{
+    sigil_err_t err;
+    int count = 0;
+    char tmp[SUBFILTER_MAX],
+            c;
+
+    sigil_zeroize(tmp, SUBFILTER_MAX * sizeof(*tmp));
+
+    err = parse_word(sgl, "/");
+    if (err != ERR_NO)
+        return err;
+
+    while ((err = pdf_peek_char(sgl, &c)) == ERR_NO) {
+        if (is_whitespace(c)) {
+            if (count <= 0)
+                return ERR_PDF_CONTENT;
+            break;
+        } else {
+            if (count >= SUBFILTER_MAX - 1)
+                return ERR_PDF_CONTENT;
+            tmp[count++] = c;
+        }
+
+        if ((err = pdf_move_pos_rel(sgl, 1)) != ERR_NO)
+            return err;
+    }
+
+    if (err != ERR_NO)
+        return err;
+
+    if (strncmp(tmp, "adbe.x509.rsa_sha1", 18) == 0) {
+        sgl->subfilter = SUBFILTER_adbe_x509_rsa_sha1;
+    } else {
+        sgl->subfilter = SUBFILTER_UNKNOWN;
+    }
+
+    return ERR_NO;
+}
+
+static sigil_err_t parse_byte_range(sigil_t *sgl)
+{