ReFuel/freetype
1
0
Fork 0
Code Issues Pull Requests Packages Projects Releases Wiki Activity
8f43c714a5
freetype/src/type1z/t1gload.c
David Turner 2dfb5aca28 added a new experimental Type 1 driver. This version sports several 
improvements compared to "src/type1". Briefly :

  - the postscript-like interpreter is now gone. We now perform
    pattern-matching to load our font content. The resultant code
    is much faster, smaller, and easier to maintain.

    For a more detailed description, see the comments at the start
    of "src/type1z/t1objs.c"

  - the glyph loader will be greatly simplified really soon, and the
    clumsy hinter will be ditched in favor of a more general auto-hinting
    module (when we get one).

    The goal is to strip down the Type1 driver to about 15-20 Kb.

This is still an experimental version. It is added to the repository
for archiving. You should not try to use it for now (it won't work
because some crucial parts are _knowingly_ ignored for now !).

Note that the "old" type1 driver will probably move to the "demos/src"
directory, in order to show how one can replace one given font driver
at runtime (and to provide the functionality of the "t1dump" program).
2000-01-27 14:02:04 +00:00

1583 lines
50 KiB
C
Raw Blame History

/*******************************************************************
*
* t1gload.c 1.0
*
* Type1 Glyph Loader.
*
* Copyright 1996-1999 by
* David Turner, Robert Wilhelm, and Werner Lemberg.
*
* This file is part of the FreeType project, and may only be used
* modified and distributed under the terms of the FreeType project
* license, LICENSE.TXT. By continuing to use, modify, or distribute
* this file you indicate that you have read the license and
* understand and accept it fully.
*
******************************************************************/
#include <t1gload.h>
#include <ftdebug.h>
#include <t1encode.h>
#include <ftstream.h>
#ifndef T1_CONFIG_OPTION_DISABLE_HINTER
#include <t1hinter.h>
#endif
/**********************************************************************/
/**********************************************************************/
/**********************************************************************/
/********** *********/
/********** *********/
/********** GENERIC CHARSTRINGS PARSING *********/
/********** *********/
/********** *********/
/**********************************************************************/
/**********************************************************************/
/**********************************************************************/
/*********************************************************************
*
* <Function>
* T1_Init_Builder
*
* <Description>
* Initialise a given glyph builder.
*
* <Input>
* builder :: glyph builder to initialise
* face :: current face object
* size :: current size object
* glyph :: current glyph object
* funcs :: glyph builder functions (or "methods").
*
*********************************************************************/
EXPORT_FUNC
void T1_Init_Builder( T1_Builder* builder,
T1_Face face,
T1_Size size,
T1_GlyphSlot glyph,
const T1_Builder_Funcs* funcs )
{
builder->funcs = *funcs;
builder->path_begun = 0;
builder->load_points = 1;
builder->face = face;
builder->size = size;
builder->glyph = glyph;
builder->memory = face->root.memory;
if (glyph)
{
builder->base = glyph->root.outline;
builder->max_points = glyph->max_points;
builder->max_contours = glyph->max_contours;
}
if (size)
{
builder->scale_x = size->root.metrics.x_scale;
builder->scale_y = size->root.metrics.y_scale;
}
builder->pos_x = 0;
builder->pos_y = 0;
builder->left_bearing.x = 0;
builder->left_bearing.y = 0;
builder->advance.x = 0;
builder->advance.y = 0;
builder->base.n_points = 0;
builder->base.n_contours = 0;
builder->current = builder->base;
builder->pass = 0;
builder->hint_point = 0;
}
/*********************************************************************
*
* <Function>
* T1_Done_Builder
*
* <Description>
* Finalise a given glyph builder. Its content can still be
* used after the call, but the function saves important information
* within the corresponding glyph slot.
*
* <Input>
* builder :: glyph builder to initialise
*
*********************************************************************/
EXPORT_FUNC
void T1_Done_Builder( T1_Builder* builder )
{
T1_GlyphSlot glyph = builder->glyph;
if (glyph)
{
glyph->root.outline = builder->base;
glyph->max_points = builder->max_points;
glyph->max_contours = builder->max_contours;
}
}
/*********************************************************************
*
* <Function>
* T1_Init_Decoder
*
* <Description>
* Initialise a given Type 1 decoder for parsing
*
* <Input>
* decoder :: Type 1 decoder to initialise
* funcs :: hinter functions interface
*
*********************************************************************/
EXPORT_FUNC
void T1_Init_Decoder( T1_Decoder* decoder,
const T1_Hinter_Funcs* funcs )
{
decoder->hinter = *funcs; /* copy hinter interface */
decoder->top = 0;
decoder->zone = 0;
decoder->flex_state = 0;
decoder->num_flex_vectors = 0;
/* Clear loader */
MEM_Set( &decoder->builder, 0, sizeof(decoder->builder) );
}
/*********************************************************************
*
* <Function>
* lookup_glyph_by_stdcharcode
*
* <Description>
* Lookup a given glyph by its StandardEncoding charcode. Used
* to implement the SEAC Type 1 operator.
*
* <Input>
* face :: current face object
* charcode :: charcode to look for
*
* <Return>
* glyph index in font face. Returns -1 if the corresponding
* glyph wasn't found.
*
*********************************************************************/
static
T1_Int lookup_glyph_by_stdcharcode( T1_Face face,
T1_Int charcode )
{
T1_Int n;
const T1_String* glyph_name;
/* check range of standard char code */
if (charcode < 0 || charcode > 255)
return -1;
glyph_name = t1_standard_strings[t1_standard_encoding[charcode]];
for ( n = 0; n < face->type1.num_glyphs; n++ )
{
T1_String* name = (T1_String*)face->type1.glyph_names[n];
if ( name && strcmp(name,glyph_name) == 0 )
return n;
}
return -1;
}
/*********************************************************************
*
* <Function>
* t1operator_seac
*
* <Description>
* Implements the "seac" Type 1 operator for a Type 1 decoder
*
* <Input>
* decoder :: current Type 1 decoder
* asb :: accent's side bearing
* adx :: horizontal position of accent
* ady :: vertical position of accent
* bchar :: base character's StandardEncoding charcode
* achar :: accent character's StandardEncoding charcode
*
* <Return>
* Error code. 0 means success.
*
*********************************************************************/
static
T1_Error t1operator_seac( T1_Decoder* decoder,
T1_Pos asb,
T1_Pos adx,
T1_Pos ady,
T1_Int bchar,
T1_Int achar )
{
T1_Error error;
T1_Face face = decoder->builder.face;
T1_Int bchar_index, achar_index, n_base_points;
FT_Outline* cur = &decoder->builder.current;
FT_Outline* base = &decoder->builder.base;
T1_Vector left_bearing, advance;
T1_Font* type1 = &face->type1;
bchar_index = lookup_glyph_by_stdcharcode( face, bchar );
achar_index = lookup_glyph_by_stdcharcode( face, achar );
if (bchar_index < 0 || achar_index < 0)
{
FT_ERROR(( "T1.Parse_Seac : invalid seac character code arguments\n" ));
return T1_Err_Syntax_Error;
}
/* First load "bchar" in builder */
/* now load the unscaled outline */
cur->n_points = 0;
cur->n_contours = 0;
cur->points = base->points + base->n_points;
cur->flags = base->flags + base->n_points;
cur->contours = base->contours + base->n_contours;
error = T1_Parse_CharStrings( decoder,
type1->charstrings [bchar_index],
type1->charstrings_len[bchar_index],
type1->num_subrs,
type1->subrs,
type1->subrs_len );
if (error) return error;
n_base_points = cur->n_points;
/* save the left bearing and width of the base character */
/* as they will be erase by the next load.. */
left_bearing = decoder->builder.left_bearing;
advance = decoder->builder.advance;
decoder->builder.left_bearing.x = 0;
decoder->builder.left_bearing.y = 0;
/* Now load "achar" on top of */
/* the base outline */
/* */
cur->n_points = 0;
cur->n_contours = 0;
cur->points = base->points + base->n_points;
cur->flags = base->flags + base->n_points;
cur->contours = base->contours + base->n_contours;
error = T1_Parse_CharStrings( decoder,
type1->charstrings [achar_index],
type1->charstrings_len[achar_index],
type1->num_subrs,
type1->subrs,
type1->subrs_len );
if (error) return error;
/* adjust contours in accented character outline */
{
T1_Int n;
for ( n = 0; n < cur->n_contours; n++ )
cur->contours[n] += n_base_points;
}
/* restore the left side bearing and */
/* advance width of the base character */
decoder->builder.left_bearing = left_bearing;
decoder->builder.advance = advance;
/* Finally, move the accent */
FT_Translate_Outline( cur, adx - asb, ady );
(void)asb; /* ignore this parameter */
return T1_Err_Ok;
}
/*********************************************************************
*
* <Function>
* t1operator_flex
*
* <Description>
* Implements the "flex" Type 1 operator for a Type 1 decoder
*
* <Input>
* decoder :: current Type 1 decoder
* threshold :: threshold
* end_x :: position of final flex point
* end_y :: position of final flex point
*
* <Return>
* Error code. 0 means success.
*
*********************************************************************/
static
T1_Error t1operator_flex( T1_Decoder* decoder,
T1_Pos threshold,
T1_Pos end_x,
T1_Pos end_y )
{
T1_Vector vec;
T1_Vector* flex = decoder->flex_vectors;
T1_Int n;
/* we don't even try to test the threshold in the non-hinting */
/* builder, even if the flex operator is said to be a path */
/* construction statement in the specification. This is better */
/* left to the hinter.. */
flex = decoder->flex_vectors;
vec = *flex++;
for ( n = 0; n < 6; n++ )
{
flex->x += vec.x;
flex->y += vec.y;
vec = *flex++;
}
(void)threshold;
(void)end_x;
(void)end_y;
flex = decoder->flex_vectors;
return decoder->builder.funcs.rcurve_to( &decoder->builder,
flex[0].x, flex[0].y,
flex[1].x, flex[1].y,
flex[2].x, flex[2].y ) ||
decoder->builder.funcs.rcurve_to( &decoder->builder,
flex[3].x, flex[3].y,
flex[4].x, flex[4].y,
flex[5].x, flex[5].y );
}
/*********************************************************************
*
* <Function>
* T1_Parse_CharStrings
*
* <Description>
* Parses a given Type 1 charstrings program
*
* <Input>
* decoder :: current Type 1 decoder
* charstring_base :: base of the charstring stream
* charstring_len :: length in bytes of the charstring stream
* num_subrs :: number of sub-routines
* subrs_base :: array of sub-routines addresses
* subrs_len :: array of sub-routines lengths
*
* <Return>
* Error code. 0 means success.
*
*********************************************************************/
EXPORT_FUNC
T1_Error T1_Parse_CharStrings( T1_Decoder* decoder,
T1_Byte* charstring_base,
T1_Int charstring_len,
T1_Int num_subrs,
T1_Byte** subrs_base,
T1_Int* subrs_len )
{
T1_Error error;
T1_Decoder_Zone* zone;
T1_Byte* ip;
T1_Byte* limit;
T1_Builder* builder = &decoder->builder;
T1_Builder_Funcs* builds = &builder->funcs;
T1_Hinter_Funcs* hints = &decoder->hinter;
static const T1_Int args_count[ op_max ] =
{
0, /* none */
0, /* endchar */
2, /* hsbw */
5, /* seac */
4, /* sbw */
0, /* closepath */
1, /* hlineto */
1, /* hmoveto */
4, /* hvcurveto */
2, /* rlineto */
2, /* rmoveto */
6, /* rrcurveto */
4, /* vhcurveto */
1, /* vlineto */
1, /* vmoveto */
0, /* dotsection */
2, /* hstem */
6, /* hstem3 */
2, /* vstem */
6, /* vstem3 */
2, /* div */
-1, /* callothersubr */
1, /* callsubr */
0, /* pop */
0, /* return */
2 /* setcurrentpoint */
};
/* First of all, initialise the decoder */
decoder->top = decoder->stack;
decoder->zone = decoder->zones;
zone = decoder->zones;
builder->path_begun = 0;
zone->base = charstring_base;
limit = zone->limit = charstring_base + charstring_len;
ip = zone->cursor = zone->base;
error = T1_Err_Ok;
/* now, execute loop */
while ( ip < limit )
{
T1_Int* top = decoder->top;
T1_Operator op = op_none;
T1_Long value = 0;
/* First of all, decompress operator or value */
switch (*ip++)
{
case 1: op = op_hstem; break;
case 3: op = op_vstem; break;
case 4: op = op_vmoveto; break;
case 5: op = op_rlineto; break;
case 6: op = op_hlineto; break;
case 7: op = op_vlineto; break;
case 8: op = op_rrcurveto; break;
case 9: op = op_closepath; break;
case 10: op = op_callsubr; break;
case 11: op = op_return; break;
case 13: op = op_hsbw; break;
case 14: op = op_endchar; break;
case 21: op = op_rmoveto; break;
case 22: op = op_hmoveto; break;
case 30: op = op_vhcurveto; break;
case 31: op = op_hvcurveto; break;
case 12:
{
if (ip > limit)
{
FT_ERROR(( "T1.Parse_CharStrings : invalid escape (12+EOF)\n" ));
goto Syntax_Error;
}
switch (*ip++)
{
case 0: op = op_dotsection; break;
case 1: op = op_vstem3; break;
case 2: op = op_hstem3; break;
case 6: op = op_seac; break;
case 7: op = op_sbw; break;
case 12: op = op_div; break;
case 16: op = op_callothersubr; break;
case 17: op = op_pop; break;
case 33: op = op_setcurrentpoint; break;
default:
FT_ERROR(( "T1.Parse_CharStrings : invalid escape (12+%d)\n",
ip[-1] ));
goto Syntax_Error;
}
}
break;
case 255: /* four bytes integer */
{
if (ip+4 > limit)
{
FT_ERROR(( "T1.Parse_CharStrings : unexpected EOF in integer\n" ));
goto Syntax_Error;
}
value = ((long)ip[0] << 24) |
((long)ip[1] << 16) |
((long)ip[2] << 8) |
ip[3];
ip += 4;
}
break;
default:
if (ip[-1] >= 32)
{
if (ip[-1] < 247)
value = (long)ip[-1] - 139;
else
{
if (++ip > limit)
{
FT_ERROR(( "T1.Parse_CharStrings : unexpected EOF in integer\n" ));
goto Syntax_Error;
}
if (ip[-2] < 251)
value = ((long)(ip[-2]-247) << 8) + ip[-1] + 108;
else
value = -((((long)ip[-2]-251) << 8) + ip[-1] + 108 );
}
}
else
{
FT_ERROR(( "T1.Parse_CharStrings : invalid byte (%d)\n",
ip[-1] ));
goto Syntax_Error;
}
}
/* push value if needed */
if ( op == op_none )
{
if ( top - decoder->stack >= T1_MAX_CHARSTRINGS_OPERANDS )
{
FT_ERROR(( "T1.Parse_CharStrings : Stack overflow !!\n" ));
goto Syntax_Error;
}
*top++ = value;
decoder->top = top;
}
else if ( op == op_callothersubr ) /* check arguments differently */
{
if ( top - decoder->stack < 2)
goto Stack_Underflow;
top -= 2;
switch (top[1])
{
case 1: /* start flex feature ----------------------------- */
{
if (top[0] != 0) goto Unexpected_OtherSubr;
decoder->flex_state = 1;
decoder->num_flex_vectors = 0;
decoder->flex_vectors[0].x = 0;
decoder->flex_vectors[0].y = 0;
}
break;
case 2: /* add flex vector ------------------------------- */
{
T1_Int index;
T1_Vector* flex;
if (top[0] != 0) goto Unexpected_OtherSubr;
top -= 2;
if (top < decoder->stack) goto Stack_Underflow;
index = decoder->num_flex_vectors++;
if (index >= 7)
{
FT_ERROR(( "T1.Parse_CharStrings: too many flex vectors !\n" ));
goto Syntax_Error;
}
flex = decoder->flex_vectors + index;
flex->x += top[0];
flex->y += top[1];
}
break;
case 0: /* end flex feature ------------------------------ */
{
if ( decoder->flex_state == 0 ||
decoder->num_flex_vectors != 7 )
{
FT_ERROR(( "T1.Parse_CharStrings: unexpected flex end\n" ));
goto Syntax_Error;
}
if (top[0] != 3) goto Unexpected_OtherSubr;
top -= 3;
if (top < decoder->stack) goto Stack_Underflow;
/* now consume the remaining "pop pop setcurrentpoint" */
if ( ip+6 > limit ||
ip[0] != 12 || ip[1] != 17 || /* pop */
ip[2] != 12 || ip[3] != 17 || /* pop */
ip[4] != 12 || ip[5] != 33 ) /* setcurrentpoint */
{
FT_ERROR(( "T1.Parse_CharStrings: invalid flex charstring\n" ));
goto Syntax_Error;
}
decoder->flex_state = 0;
decoder->top = top;
error = t1operator_flex( decoder, top[0], top[1], top[2] );
}
break;
case 3: /* change hints ------------------------------------ */
{
if (top[0] != 1) goto Unexpected_OtherSubr;
/* eat the following "pop" */
if (ip+2 > limit)
{
FT_ERROR(( "T1.Parse_CharStrings: invalid escape (12+%d)\n",
ip[-1] ));
goto Syntax_Error;
}
if (ip[0] != 12 || ip[1] != 17)
{
FT_ERROR(( "T1.Parse_CharStrings: 'pop' expected, found (%d %d)\n",
ip[0], ip[1] ));
goto Syntax_Error;
}
ip += 2;
error = hints->change_hints(builder);
}
break;
default:
/* invalid OtherSubrs call */
Unexpected_OtherSubr:
FT_ERROR(( "T1.Parse_CharStrings: unexpected OtherSubrs [%d %d]\n",
top[0], top[1] ));
goto Syntax_Error;
}
decoder->top = top;
}
else
{
T1_Int num_args = args_count[op];
if ( top - decoder->stack < num_args )
goto Stack_Underflow;
top -= num_args;
switch (op)
{
case op_endchar:
error = builds->end_char( builder );
break;
case op_hsbw:
error = builds->set_bearing_point( builder, top[0], 0,
top[1], 0 );
break;
case op_seac:
/* return immediately after the processing */
return t1operator_seac( decoder, top[0], top[1],
top[2], top[3], top[4] );
case op_sbw:
error = builds->set_bearing_point( builder, top[0], top[1],
top[2], top[3] );
break;
case op_closepath:
error = builds->close_path( builder );
break;
case op_hlineto:
error = builds->rline_to( builder, top[0], 0 );
break;
case op_hmoveto:
error = builds->rmove_to( builder, top[0], 0 );
break;
case op_hvcurveto:
error = builds->rcurve_to( builder, top[0], 0,
top[1], top[2],
0, top[3] );
break;
case op_rlineto:
error = builds->rline_to( builder, top[0], top[1] );
break;
case op_rmoveto:
/* ignore operator when in flex mode */
if (decoder->flex_state == 0)
error = builds->rmove_to( builder, top[0], top[1] );
else
top += 2;
break;
case op_rrcurveto:
{
error = builds->rcurve_to( builder, top[0], top[1],
top[2], top[3],
top[4], top[5] );
}
break;
case op_vhcurveto:
error = builds->rcurve_to( builder, 0, top[0],
top[1], top[2],
top[3], 0 );
break;
case op_vlineto:
error = builds->rline_to( builder, 0, top[0] );
break;
case op_vmoveto:
error = builds->rmove_to( builder, 0, top[0] );
break;
case op_dotsection:
error = hints->dot_section( builder );
break;
case op_hstem:
error = hints->stem( builder, top[0], top[1], 0 );
break;
case op_hstem3:
error = hints->stem3( builder, top[0], top[1], top[2],
top[3], top[4], top[5], 0 );
break;
case op_vstem:
error = hints->stem( builder, top[0], top[1], 1 );
break;
case op_vstem3:
error = hints->stem3( builder, top[0], top[1], top[2],
top[3], top[4], top[5], 1 );
break;
case op_div:
if (top[1])
*top++ = top[0] / top[1];
else
{
FT_ERROR(( "T1.Parse_CHarStrings : division by 0\n" ));
goto Syntax_Error;
}
break;
case op_callsubr:
{
T1_Int index = top[0];
if ( index < 0 || index >= num_subrs )
{
FT_ERROR(( "T1.Parse_CharStrings : invalid subrs index\n" ));
goto Syntax_Error;
}
if ( zone - decoder->zones >= T1_MAX_SUBRS_CALLS )
{
FT_ERROR(( "T1.Parse_CharStrings : too many nested subrs\n" ));
goto Syntax_Error;
}
zone->cursor = ip; /* save current instruction pointer */
zone++;
zone->base = subrs_base[index];
zone->limit = zone->base + subrs_len[index];
zone->cursor = zone->base;
if (!zone->base)
{
FT_ERROR(( "T1.Parse_CharStrings : invoking empty subrs !!\n" ));
goto Syntax_Error;
}
decoder->zone = zone;
ip = zone->base;
limit = zone->limit;
}
break;
case op_pop:
FT_ERROR(( "T1.Parse_CharStrings : unexpected POP\n" ));
goto Syntax_Error;
case op_return:
if ( zone <= decoder->zones )
{
FT_ERROR(( "T1.Parse_CharStrings : unexpected return\n" ));
goto Syntax_Error;
}
zone--;
ip = zone->cursor;
limit = zone->limit;
decoder->zone = zone;
break;
case op_setcurrentpoint:
FT_ERROR(( "T1.Parse_CharStrings : unexpected SETCURRENTPOINT\n" ));
goto Syntax_Error;
break;
default:
FT_ERROR(( "T1.Parse_CharStrings : unhandled opcode %d\n", op ));
goto Syntax_Error;
}
decoder->top = top;
}
}
return error;
Syntax_Error:
return T1_Err_Syntax_Error;
Stack_Underflow:
return T1_Err_Stack_Underflow;
}
/*************************************************************************/
/* */
/* <Function> T1_Add_Points */
/* */
/* <Description> */
/* Checks that there is enough room in the current load glyph outline */
/* to accept "num_points" additional outline points. If not, this */
/* function grows the load outline's arrays accordingly.. */
/* */
/* <Input> */
/* builder :: pointer to glyph builder object */
/* num_points :: number of points that will be added later */
/* */
/* <Return> */
/* Type1 error code. 0 means success */
/* */
/* <Note> */
/* This function does NOT update the points count in the glyph builder*/
/* This must be done by the caller itself, after this function is */
/* invoked.. */
/* */
LOCAL_FUNC
T1_Error T1_Add_Points( T1_Builder* builder,
T1_Int num_points )
{
T1_Int new_points;
new_points = builder->base.n_points +
builder->current.n_points +
num_points;
if ( new_points > builder->max_points )
{
FT_Memory memory = builder->memory;
T1_Error error;
T1_Int increment = builder->current.points - builder->base.points;
T1_Int current = builder->max_points;
while ( builder->max_points < new_points )
builder->max_points += 16;
if ( REALLOC_ARRAY( builder->base.points,
current, builder->max_points, T1_Vector ) ||
REALLOC_ARRAY( builder->base.flags,
current, builder->max_points, T1_Byte ) )
return error;
builder->current.points = builder->base.points + increment;
builder->current.flags = builder->base.flags + increment;
}
return T1_Err_Ok;
}
/*************************************************************************/
/* */
/* <Function> T1_Add_Contours */
/* */
/* <Description> */
/* Checks that there is enough room in the current load glyph outline */
/* to accept "num_contours" additional contours. If not, this func */
/* the load outline's arrays accordingly.. */
/* */
/* <Input> */
/* builder :: pointer to glyph builder object */
/* num_contours :: number of contours that will be added later */
/* */
/* <Return> */
/* Type1 error code. 0 means success */
/* */
/* <Note> */
/* This function does NOT update the contours count in the load glyph */
/* This must be done by the caller itself, after this function is */
/* invoked.. */
/* */
LOCAL_FUNC
T1_Error T1_Add_Contours( T1_Builder* builder,
T1_Int num_contours )
{
T1_Int new_contours;
new_contours = builder->base.n_contours +
builder->current.n_contours +
num_contours;
if ( new_contours > builder->max_contours && builder->load_points )
{
T1_Error error;
FT_Memory memory = builder->memory;
T1_Int increment = builder->current.contours - builder->base.contours;
T1_Int current = builder->max_contours;
while ( builder->max_contours < new_contours )
builder->max_contours += 4;
if ( REALLOC_ARRAY( builder->base.contours,
current, builder->max_contours, T1_Short ) )
return error;
builder->current.contours = builder->base.contours + increment;
}
return T1_Err_Ok;
}
/**********************************************************************/
/**********************************************************************/
/**********************************************************************/
/********** *********/
/********** *********/
/********** COMPUTE THE MAXIMUM ADVANCE WIDTH *********/
/********** *********/
/********** The following code is in charge of computing *********/
/********** the maximum advance width of the font. It *********/
/********** quickly process each glyph charstring to *********/
/********** extract the value from either a "sbw" or "seac" *********/
/********** operator. *********/
/********** *********/
/**********************************************************************/
/**********************************************************************/
/**********************************************************************/
static
T1_Error maxadv_sbw( T1_Decoder* decoder,
T1_Pos sbx,
T1_Pos sby,
T1_Pos wx,
T1_Pos wy )
{
if (wx > decoder->builder.advance.x)
decoder->builder.advance.x = wx;
(void)sbx;
(void)sby;
(void)wy;
return -1; /* return an error code to exit the Type 1 parser */
/* immediately. */
}
static
T1_Int maxadv_error( void )
{
/* we should never reach this code, unless with a buggy font */
return -2;
}
/* the maxadv_gbuilder_interface is used when computing the maximum */
/* advance width of all glyphs in a given font. We only process the */
/* 'sbw' operator here, and return an error for all others.. */
/* Note that "seac" is processed by the T1_Decoder */
static
const T1_Builder_Funcs maxadv_builder_interface =
{
(T1_Builder_EndChar) maxadv_error,
(T1_Builder_Sbw) maxadv_sbw,
(T1_Builder_ClosePath) maxadv_error,
(T1_Builder_RLineTo) maxadv_error,
(T1_Builder_RMoveTo) maxadv_error,
(T1_Builder_RCurveTo) maxadv_error
};
/* the maxadv_interface is used when computing the maximum advance */
/* with of the set of glyphs in a given font file. We only process */
/* the "seac" operator and return immediately.. */
static
const T1_Hinter_Funcs maxadv_hinter_interface =
{
(T1_Hinter_DotSection) maxadv_error,
(T1_Hinter_ChangeHints) maxadv_error,
(T1_Hinter_Stem) maxadv_error,
(T1_Hinter_Stem3) maxadv_error,
};
LOCAL_FUNC
T1_Error T1_Compute_Max_Advance( T1_Face face,
T1_Int *max_advance )
{
T1_Error error;
T1_Decoder decoder;
T1_Int glyph_index;
T1_Font* type1 = &face->type1;
*max_advance = 0;
/* Initialise load decoder */
T1_Init_Decoder( &decoder, &maxadv_hinter_interface );
T1_Init_Builder( &decoder.builder, face, 0, 0,
&maxadv_builder_interface );
/* For each glyph, parse the glyph charstring and extract */
/* the advance width.. */
for ( glyph_index = 0; glyph_index < type1->num_glyphs; glyph_index++ )
{
/* now get load the unscaled outline */
error = T1_Parse_CharStrings( &decoder,
type1->charstrings [glyph_index],
type1->charstrings_len[glyph_index],
type1->num_subrs,
type1->subrs,
type1->subrs_len );
/* ignore the error if one occured - skip to next glyph */
(void)error;
}
*max_advance = decoder.builder.advance.x;
return T1_Err_Ok;
}
/**********************************************************************/
/**********************************************************************/
/**********************************************************************/
/********** *********/
/********** *********/
/********** UNHINTED GLYPH LOADER *********/
/********** *********/
/********** The following code is in charge of loading a *********/
/********** single outline. It completely ignores hinting *********/
/********** and is used when FT_LOAD_NO_HINTING is set. *********/
/********** *********/
/********** The Type 1 hinter is located in "t1hint.c" *********/
/********** *********/
/**********************************************************************/
/**********************************************************************/
/**********************************************************************/
static
T1_Error close_open_path( T1_Builder* builder )
{
T1_Error error;
FT_Outline* cur = &builder->current;
T1_Int num_points;
T1_Int first_point;
/* Some fonts, like Hershey, are made of "open paths" which are */
/* now managed directly by FreeType. In this case, it is necessary */
/* to close the path by duplicating its points in reverse order, */
/* which is precisely the purpose of this function */
/* first compute the number of points to duplicate.. */
if (cur->n_contours > 1)
first_point = cur->contours[ cur->n_contours-2 ]+1;
else
first_point = 0;
num_points = cur->n_points - first_point - 2;
if ( num_points > 0 )
{
T1_Vector* source_point;
char* source_flags;
T1_Vector* point;
char* flags;
error = T1_Add_Points( builder, num_points );
if (error) return error;
point = cur->points + cur->n_points;
flags = cur->flags + cur->n_points;
source_point = point - 2;
source_flags = flags - 2;
cur->n_points += num_points;
if ( builder->load_points )
do
{
*point++ = *source_point--;
*flags++ = *source_flags--;
num_points--;
}
while (num_points > 0);
}
builder->path_begun = 0;
return T1_Err_Ok;
}
static
T1_Error gload_closepath( T1_Builder* builder )
{
FT_Outline* cur = &builder->current;
/* save current contour, if any */
if ( cur->n_contours > 0 )
cur->contours[cur->n_contours-1] = cur->n_points-1;
#ifndef T1_CONFIG_OPTION_DISABLE_HINTER
/* hint latest points if needed - this is not strictly required */
/* there, but it helps for debugging, and doesn't affect performance */
if ( builder->pass == 1 )
T1_Hint_Points( builder );
#endif
builder->path_begun = 0;
return T1_Err_Ok;
}
static
T1_Error gload_endchar( T1_Builder* builder )
{
FT_Outline* cur = &builder->current;
T1_Error error;
/* close path if needed */
if (builder->path_begun)
{
error = close_open_path( builder );
if (error) return error;
}
error = gload_closepath( builder );
builder->base.n_points += cur->n_points;
builder->base.n_contours += cur->n_contours;
return error;
}
static
T1_Error gload_sbw( T1_Builder* builder,
T1_Pos sbx,
T1_Pos sby,
T1_Pos wx,
T1_Pos wy )
{
builder->left_bearing.x += sbx;
builder->left_bearing.y += sby;
builder->advance.x = wx;
builder->advance.y = wy;
builder->last.x = sbx;
builder->last.y = sby;
return 0;
}
static
T1_Error gload_rlineto( T1_Builder* builder,
T1_Pos dx,
T1_Pos dy )
{
T1_Error error;
FT_Outline* cur = &builder->current;
T1_Vector vec;
/* grow buffer if necessary */
error = T1_Add_Points ( builder, 1 );
if (error) return error;
if ( builder->load_points )
{
/* save point */
vec.x = builder->last.x + dx;
vec.y = builder->last.y + dy;
cur->points[cur->n_points] = vec;
cur->flags [cur->n_points] = FT_Curve_Tag_On;
builder->last = vec;
}
cur->n_points++;
builder->path_begun = 1;
return T1_Err_Ok;
}
static
T1_Error gload_rmoveto( T1_Builder* builder,
T1_Pos dx,
T1_Pos dy )
{
T1_Error error;
FT_Outline* cur = &builder->current;
T1_Vector vec;
/* in the case where "path_begun" is set, we have a rmoveto */
/* after some normal path definition. When the face's paint */
/* type is set to 1, this means that we have an "open path", */
/* also called a 'stroke'. The FreeType raster doesn't support */
/* opened path, so we'll close it explicitely there.. */
if ( builder->path_begun && builder->face->type1.paint_type == 1 )
{
if ( builder->face->type1.paint_type == 1 )
{
error = close_open_path( builder );
if (error) return error;
}
}
/* grow buffer if necessary */
error = T1_Add_Contours( builder, 1 ) ||
T1_Add_Points ( builder, 1 );
if (error) return error;
/* save current contour, if any */
if ( cur->n_contours > 0 )
cur->contours[cur->n_contours-1] = cur->n_points-1;
if ( builder->load_points )
{
/* save point */
vec.x = builder->last.x + dx;
vec.y = builder->last.y + dy;
cur->points[cur->n_points] = vec;
cur->flags [cur->n_points] = FT_Curve_Tag_On;
builder->last = vec;
}
cur->n_contours++;
cur->n_points++;
return T1_Err_Ok;
}
static
T1_Error gload_rrcurveto( T1_Builder* builder,
T1_Pos dx1,
T1_Pos dy1,
T1_Pos dx2,
T1_Pos dy2,
T1_Pos dx3,
T1_Pos dy3 )
{
T1_Error error;
FT_Outline* cur = &builder->current;
T1_Vector vec;
T1_Vector* points;
char* flags;
/* grow buffer if necessary */
error = T1_Add_Points ( builder, 3 );
if (error) return error;
if ( builder->load_points )
{
/* save point */
points = cur->points + cur->n_points;
flags = cur->flags + cur->n_points;
vec.x = builder->last.x + dx1;
vec.y = builder->last.y + dy1;
points[0] = vec; flags[0] = FT_Curve_Tag_Cubic;
vec.x += dx2;
vec.y += dy2;
points[1] = vec; flags[1] = FT_Curve_Tag_Cubic;
vec.x += dx3;
vec.y += dy3;
points[2] = vec; flags[2] = FT_Curve_Tag_On;
builder->last = vec;
}
cur->n_points += 3;
builder->path_begun = 1;
return T1_Err_Ok;
}
static
T1_Error gload_ignore( void )
{
return 0;
}
static
const T1_Builder_Funcs gload_builder_interface =
{
gload_endchar,
gload_sbw,
gload_closepath,
gload_rlineto,
gload_rmoveto,
gload_rrcurveto
};
static
const T1_Builder_Funcs gload_builder_interface_null =
{
(T1_Builder_EndChar) gload_ignore,
(T1_Builder_Sbw) gload_sbw, /* record left bearing */
(T1_Builder_ClosePath) gload_ignore,
(T1_Builder_RLineTo) gload_ignore,
(T1_Builder_RMoveTo) gload_ignore,
(T1_Builder_RCurveTo) gload_ignore
};
static
const T1_Hinter_Funcs gload_hinter_interface =
{
(T1_Hinter_DotSection) gload_ignore, /* dotsection */
(T1_Hinter_ChangeHints) gload_ignore, /* changehints */
(T1_Hinter_Stem) gload_ignore, /* hstem & vstem */
(T1_Hinter_Stem3) gload_ignore, /* hstem3 & vestem3 */
};
LOCAL_FUNC
T1_Error T1_Load_Glyph( T1_GlyphSlot glyph,
T1_Size size,
T1_Int glyph_index,
T1_Int load_flags )
{
T1_Error error;
T1_Decoder decoder;
T1_Face face = (T1_Face)glyph->root.face;
T1_Bool hinting;
T1_Font* type1 = &face->type1;
glyph->x_scale = size->root.metrics.x_scale;
glyph->y_scale = size->root.metrics.y_scale;
glyph->root.outline.n_points = 0;
glyph->root.outline.n_contours = 0;
hinting = ( load_flags & FT_LOAD_NO_SCALE ) == 0 &&
( load_flags & FT_LOAD_NO_HINTING ) == 0;
glyph->root.format = ft_glyph_format_none;
#ifndef T1_CONFIG_OPTION_DISABLE_HINTER
/*****************************************************************/
/* */
/* Hinter overview : */
/* */
/* This is a two-pass hinter. On the first pass, the hints */
/* are all recorded by the hinter, and no point is loaded */
/* in the outline. */
/* */
/* When the first pass is finished, all stems hints are */
/* grid-fitted at once. */
/* */
/* Then, a second pass is performed to load the outline */
/* points as well as hint/scale them correctly. */
/* */
if ( hinting )
{
/* Pass 1 - don't record points, simply stem hints */
T1_Init_Decoder( &decoder, &t1_hinter_funcs );
T1_Init_Builder( &decoder.builder, face, size, glyph,
&gload_builder_interface_null );
glyph->hints->hori_stems.num_stems = 0;
glyph->hints->vert_stems.num_stems = 0;
error = T1_Parse_CharStrings( &decoder,
type1->charstrings [glyph_index],
type1->charstrings_len[glyph_index],
type1->num_subrs,
type1->subrs,
type1->subrs_len );
/* All right, pass 1 is finished, now grid-fit all stem hints */
T1_Hint_Stems( &decoder.builder );
/* Pass 2 - record and scale/hint the points */
T1_Init_Decoder( &decoder, &t1_hinter_funcs );
T1_Init_Builder( &decoder.builder, face, size, glyph,
&gload_builder_interface );
decoder.builder.pass = 1;
error = T1_Parse_CharStrings( &decoder,
type1->charstrings [glyph_index],
type1->charstrings_len[glyph_index],
type1->num_subrs,
type1->subrs,
type1->subrs_len );
/* save new glyph tables */
T1_Done_Builder( &decoder.builder );
}
else
#endif
{
T1_Init_Decoder( &decoder, &gload_hinter_interface );
T1_Init_Builder( &decoder.builder, face, size, glyph,
&gload_builder_interface );
/* now load the unscaled outline */
error = T1_Parse_CharStrings( &decoder,
type1->charstrings [glyph_index],
type1->charstrings_len[glyph_index],
type1->num_subrs,
type1->subrs,
type1->subrs_len );
/* save new glyph tables */
T1_Done_Builder( &decoder.builder );
}
/* Now, set the metrics.. - this is rather simple, as : */
/* the left side bearing is the xMin, and the top side */
/* bearing the yMax.. */
if (!error)
{
FT_BBox cbox;
FT_Glyph_Metrics* metrics = &glyph->root.metrics;
FT_Get_Outline_CBox( &glyph->root.outline, &cbox );
/* grid fit the bounding box if necessary */
if (hinting)
{
cbox.xMin &= -64;
cbox.yMin &= -64;
cbox.xMax = ( cbox.xMax+63 ) & -64;
cbox.yMax = ( cbox.yMax+63 ) & -64;
}
metrics->width = cbox.xMax - cbox.xMin;
metrics->height = cbox.yMax - cbox.yMin;
metrics->horiBearingX = cbox.xMin;
metrics->horiBearingY = cbox.yMax;
/* copy the _unscaled_ advance width */
metrics->horiAdvance = decoder.builder.advance.x;
/* make up vertical metrics */
metrics->vertBearingX = 0;
metrics->vertBearingY = 0;
metrics->vertAdvance = 0;
glyph->root.format = ft_glyph_format_outline;
glyph->root.outline.second_pass = TRUE;
glyph->root.outline.high_precision = ( size->root.metrics.y_ppem < 24 );
glyph->root.outline.dropout_mode = 2;
if ( hinting )
{
/* adjust the advance width */
/* XXX : TODO : consider stem hints grid-fit */
metrics->horiAdvance = FT_MulFix( metrics->horiAdvance,
glyph->x_scale );
}
else if ( (load_flags & FT_LOAD_NO_SCALE) == 0 )
{
/* scale the outline and the metrics */
T1_Int n;
FT_Outline* cur = &decoder.builder.base;
T1_Vector* vec = cur->points;
T1_Fixed x_scale = glyph->x_scale;
T1_Fixed y_scale = glyph->y_scale;
/* First of all, scale the points */
for ( n = cur->n_points; n > 0; n--, vec++ )
{
vec->x = FT_MulFix( vec->x, x_scale );
vec->y = FT_MulFix( vec->y, y_scale );
}
/* Then scale the metrics */
metrics->width = FT_MulFix( metrics->width, x_scale );
metrics->height = FT_MulFix( metrics->height, y_scale );
metrics->horiBearingX = FT_MulFix( metrics->horiBearingX, x_scale );
metrics->horiBearingY = FT_MulFix( metrics->horiBearingY, y_scale );
metrics->horiAdvance = FT_MulFix( metrics->horiAdvance, x_scale );
metrics->vertBearingX = FT_MulFix( metrics->vertBearingX, x_scale );
metrics->vertBearingY = FT_MulFix( metrics->vertBearingY, y_scale );
metrics->vertAdvance = FT_MulFix( metrics->vertAdvance, x_scale );
}
}
return error;
}
Reference in New Issue View Git Blame Copy Permalink