Figure out more parts of format.

diff --git a/dump.c b/dump.c
index 212421cb8045f5d330129c1882745abe1ae4388c..12ee73ebdf4d02d143776cb252eca089ac215a7e 100644 (file)
--- a/dump.c
+++ b/dump.c
@@ -130,6 +130,27 @@ match_u32_assert(uint32_t x, const char *where)
 }
 #define match_u32_assert(x) match_u32_assert(x, WHERE)
 
+static bool __attribute__((unused))
+match_u64(uint64_t x)
+{
+  if (get_u64() == x)
+    return true;
+  pos -= 8;
+  return false;
+}
+
+static void __attribute__((unused))
+match_u64_assert(uint64_t x, const char *where)
+{
+  unsigned long long int y = get_u64();
+  if (x != y)
+    {
+      fprintf(stderr, "%s: 0x%x: expected u64:%llu, got u64:%llu\n", where, pos - 8, x, y);
+      exit(1);
+    }
+}
+#define match_u64_assert(x) match_u64_assert(x, WHERE)
+
 static bool __attribute__((unused))
 match_be32(uint32_t x)
 {
@@ -320,18 +341,17 @@ hex_dump(int ofs, int n)
       int c = data[ofs + i];
 #if 1
       if (i && !(i % 16))
-        fprintf(stderr, "-");
+        printf("-");
       else
-        fprintf(stderr, " ");
+        printf(" ");
 #endif
-      fprintf(stderr, "%02x", c);
+      printf("%02x", c);
     }
   for (int i = 0; i < n; i++)
     {
       int c = data[ofs + i];
-      fprintf(stderr, "%c", c >= 32 && c < 127 ? c : '.');
+      printf("%c", c >= 32 && c < 127 ? c : '.');
     }
-  fprintf(stderr, "\n");
 }
 
 static char *
@@ -996,7 +1016,7 @@ dump_fonts(void)
     }
   bool grid = get_byte();
   pos += 3;
-  printf("  <grid show=\"%s\">\n", grid ? "yes" : "no");
+  printf("  <grid show=\"%s\"/>\n", grid ? "yes" : "no");
   printf("</borders>\n");
   assert(pos == x1_end);
 
@@ -1020,7 +1040,11 @@ dump_fonts(void)
       if (!get_byte())
         printf(" footnoteposition=\"subscript\"");
       get_byte();
-      pos += get_be32();
+      int nbytes = get_be32();
+      printf("\n");
+      hex_dump(pos, nbytes);
+      printf("\n");
+      pos += nbytes;
       get_string_be();
       char *look = get_string_be();
       if (look[0])
@@ -1029,8 +1053,16 @@ dump_fonts(void)
     }
   pos = x3_end;
 
+  /* Manual column widths, if present. */
   int count = get_u32();
-  pos += 4 * count;
+  printf("<columnwidths>");
+  for (int i = 0; i < count; i++)
+    {
+      if (i)
+        putchar(' ');
+      printf("%d", get_u32());
+    }
+  printf("</columnwidths>\n");
 
   const char *locale = get_string();
   printf ("<locale>%s</locale>\n", locale);
@@ -1076,7 +1108,35 @@ dump_fonts(void)
       int outer_end = get_end();
 
       /* First inner envelope: byte*33 int[n] int*[n]. */
-      pos = get_end();
+      int inner_len = get_u32();
+      int inner_end = pos + inner_len;
+      int array_start = pos + 33;
+      match_byte_assert(0);
+      pos++;                    /* 0, 1, 10 seen. */
+      match_byte_assert(0);
+      pos++;                    /* 0...11 seen. */
+      if (!match_byte(0) && !match_byte(1) && !match_byte(2))
+        match_byte_assert(3);
+      if (!match_byte(0) && !match_byte(2))
+        match_byte_assert(3);
+      if (!match_u64(0))
+        match_u64_assert(UINT64_MAX);
+      match_u32_assert(0);
+      match_u32_assert(0);
+      match_u32_assert(0);
+      match_u32_assert(0);
+      match_byte_assert(0);
+      if (!match_byte(0))
+        match_byte_assert(1);
+      match_byte_assert(1);
+      pos = array_start;
+#if 1
+      printf("widths:");
+      while (pos < inner_end)
+        printf(" %d", get_u32());
+      printf("\n");
+#endif
+      pos = inner_end;;
 
       /* Second inner envelope. */
       assert(get_end() == outer_end);
@@ -1117,9 +1177,9 @@ dump_fonts(void)
             match_byte_assert(0);
         }
 
-      pos += 8;
-      match_byte_assert(1);
+      printf ("small: %g\n", get_double());
 
+      match_byte_assert(1);
       if (outer_end - pos > 6)
         {
           /* There might be a pair of strings representing a dataset and
@@ -1158,7 +1218,7 @@ dump_fonts(void)
       else
         match_u32_assert(0);
 
-      match_byte_assert(0x2e);
+      match_byte_assert('.');
       if (!match_byte(0))
         match_byte_assert(1);
 
@@ -1301,10 +1361,17 @@ main(int argc, char *argv[])
       match_byte_assert(0);
 
       /* Offset 27. */
+#if 1
       pos++;
       pos++;
       pos++;
       pos++;
+#else
+      printf("\nheader:");
+      for (int i = 0; i < 4; i++)
+        printf(" %02x", data[pos++]);
+      printf("\n");
+#endif
 
       /* Offset 31. */
       printf("<tableid>%lld</tableid>", get_u64());

diff --git a/spv-file-format.texi b/spv-file-format.texi
index c5d327407feb24465714e5fa51b162d7ae7a624d..6c690920ce504d3c649e1fd8067bf0963dbe9716 100644 (file)
--- a/spv-file-format.texi
+++ b/spv-file-format.texi
@@ -572,7 +572,7 @@ concatenated together, terminated by a byte 01:
 LightMember @result{}
     Header Title
     Caption Footnotes
-    Fonts Formats Borders PrintSettings TableSettings
+    Fonts Borders PrintSettings TableSettings Formats
     Dimensions Data
     01
 @end format
@@ -583,7 +583,7 @@ The following sections go into more detail.
 @menu
 * SPV Light Member Header::
 * SPV Light Member Title::
-* PSV Light Member Caption::
+* SPV Light Member Caption::
 * SPV Light Member Footnotes::
 * SPV Light Member Fonts::
 * SPV Light Member Borders::
@@ -648,7 +648,7 @@ appropriate for presentation, and localized to the user's language,
 well formatted.  For example, for a frequency table, @code{title1} and
 @code{title2} name the variable and @code{c} is simply ``Frequencies''.
 
-@node PSV Light Member Caption
+@node SPV Light Member Caption
 @subsection Caption
 
 @cartouche
@@ -753,8 +753,8 @@ endianness.
 @code{show-grid-lines} is 1 to draw grid lines, otherwise 0.
 
 Each Border describes one kind of border.  @code{n-borders} seems to
-always be 19.  Each @code{border-type} appears once in order, and they
-correspond to the following borders:
+always be 19.  Each @code{border-type} appears once (although in an
+unpredictable order) and correspond to the following borders:
 
 @table @asis
 @item 0
@@ -886,32 +886,35 @@ TableSettings ends with an arbitrary number of null bytes.
 @cartouche
 @format
 Formats @result{}
-    int[@t{n4}] int*[@t{n4}]
+    int[@t{nwidths}] int*[@t{nwidths}]
     string[@t{encoding}]
-    (i0 @math{|} i-1) (00 @math{|} 01) 00 (00 @math{|} 01)
-    int
+    int (00 @math{|} 01) 00 (00 @math{|} 01)
+    int[@t{epoch}]
     byte[@t{decimal}] byte[@t{grouping}]
-    int[@t{n-ccs}] string*[@t{n-ccs}]
+    CustomCurrency
     v1(i0)
     v3(count(count(X5) count(X6)))
 
+CustomCurrency @result{} int[@t{n-ccs}] string*[@t{n-ccs}]
+
 X5 @result{} byte*33 int[@t{n}] int*[@t{n}]
 X6 @result{}
     01 00 (03 @math{|} 04) 00 00 00
     string[@t{command}] string[@t{subcommand}]
     string[@t{language}] string[@t{charset}] string[@t{locale}]
-    (00 @math{|} 01) 00 (00 @math{|} 01) (00 @math{|} 01)
+    (00 @math{|} 01) 00 bool bool
     int[@t{epoch}]
     byte[@t{decimal}] byte[@t{grouping}]
-    (2d 43 1c eb e2 36 1a 3f | 00*8) 01
+    double[@t{small}] 01
     (string[@t{dataset}] string[@t{datafile}] i0 int[@t{date}] i0)?
-    int[@t{n-ccs}] string*[@t{n-ccs}]
-    2e (00 @math{|} 01) (i2000000 i0)?
+    CustomCurrency
+    byte[@t{missing}] bool (i2000000 i0)?
 @end format
 @end cartouche
 
-Observed values of @code{n4} vary from 0 to 17.  Out of 7,060 examples
-in the corpus, it is nonzero only 36 times.
+If @code{nwidths} is nonzero, then the accompanying integers are
+column widths as manually adjusted by the user.  (Row heights are
+computed automatically based on the widths.)
 
 @code{encoding} is a character encoding, usually a Windows code page
 such as @code{en_US.windows-1252} or @code{it_IT.windows-1252}.  The