debian/cdesktop
1
0
Fork 0
Code Issues Pull Requests Actions Packages Projects Releases Wiki Activity

Initial import of the CDE 2.1.30 sources from the Open Group.

Browse Source
This commit is contained in:
Peter Howkins
2012-03-10 18:21:40 +00:00
commit 83b6996daa
18978 changed files with 3945623 additions and 0 deletions
Download Patch File Download Diff File Expand all files Collapse all files

452
cde/osf/uil/UilSymNam.c Normal file
View File

@@ -0,0 +1,452 @@
/*
* @OSF_COPYRIGHT@
* COPYRIGHT NOTICE
* Copyright (c) 1990, 1991, 1992, 1993 Open Software Foundation, Inc.
* ALL RIGHTS RESERVED (MOTIF). See the file named COPYRIGHT.MOTIF for
* the full copyright text.
*/
/*
* HISTORY
*/
#ifdef REV_INFO
#ifndef lint
static char rcsid[] = "$TOG: UilSymNam.c /main/13 1997/09/08 11:12:50 cshi $"
#endif
#endif
/*
* (c) Copyright 1989, 1990, DIGITAL EQUIPMENT CORPORATION, MAYNARD, MASS. */
/*
**++
** FACILITY:
**
** User Interface Language Compiler (UIL)
**
** ABSTRACT:
**
** This module inserts names into the name table.
**
**--
**/
/*
**
** INCLUDE FILES
**
**/
#include "UilDefI.h"
/*
**
** DEFINE and MACRO DEFINITIONS
**
**/
/*
**
** EXTERNAL VARIABLE DECLARATIONS
**
**/
/*
**
** GLOBAL VARIABLE DECLARATIONS
**
**/
/*
**
** OWN VARIABLE DECLARATIONS
**
**/
/*
**++
** FUNCTIONAL DESCRIPTION:
**
** This routine searches for a name entry of the same name as its parameters.
** If the entry is found, a pointer to that name node is
** returned as the value of the function. If no entry is found, a NULL
** pointer is returned.
**
** See sym_insert_name for a description of the name lookup alorithm.
**
** FORMAL PARAMETERS:
**
** l_length length of the name not including the null
** c_text pointer to a null terminated string for name
**
** IMPLICIT INPUTS:
**
** sym_az_hash_table the hash table
**
** IMPLICIT OUTPUTS:
**
** sym_az_hash_table may be updated with an additional name
**
** FUNCTION VALUE:
**
** a pointer to a name entry
**
** SIDE EFFECTS:
**
** none
**
**--
**/
sym_name_entry_type
*sym_find_name(l_length, c_text)
int l_length; /* length of name to find */
char *c_text; /* text of the name */
{
sym_name_entry_type *az_current_name;
int l_hash_code;
int l_compare_result;
/* obtain the hash code of for the name */
l_hash_code = hash_function( l_length, c_text );
/*
** chain along hash chain looking for symbol - exit loop under 3 condition
** 1) come to the end of the chain: name not found
** 2) find symbol: return this symbol
** 3) find node > symbol: name not found
*/
for (az_current_name = sym_az_hash_table[ l_hash_code ];
az_current_name != NULL;
az_current_name = az_current_name->az_next_name_entry)
{
l_compare_result = _compare(c_text, az_current_name->c_text);
if (l_compare_result == 0) /* c_text = current name */
{
/* found the name we are looking for */
return az_current_name;
}
if (l_compare_result > 0) /* c_text > current name */
{
/* return NULL - name should be before this spot in list */
return NULL;
}
}
/* came to end of the list without finding the name */
return NULL;
}
/*
**++
** FUNCTIONAL DESCRIPTION:
**
** This routine searches for a name entry of the same name as its parameters.
** If the entry is found, a pointer to that name node is
** returned as the value of the function. If no entry is found, one is
** inserted. In this case the value of the function is a pointer to
** the name entry created.
**
** Name entries are linked off of a hash table. Those
** entries that have the same hash code, are sorted according to the
** collating sequence. Thus the algorithm involves hashing the symbol and
** then following the chain for that hash code until one of the following
** conditions is met. 1) the identifier is found, then return a pointer
** to that name entry. 2) come to the end of the chain or a name
** entry that comes later in the collating sequence than the symbol being
** searched for. In this case the name is inserted just prior to this
** point in the chain.
**
** FORMAL PARAMETERS:
**
** l_length length of the name not including the null
** c_text pointer to a null terminated string for name
**
** IMPLICIT INPUTS:
**
** sym_az_hash_table the hash table
**
** IMPLICIT OUTPUTS:
**
** sym_az_hash_table may be updated with an additional name
**
** FUNCTION VALUE:
**
** a pointer to a name entry
**
** SIDE EFFECTS:
**
** may create a name entry and update the hash table
**
**--
**/
sym_name_entry_type *sym_insert_name(l_length, c_text)
int l_length; /* length of name to insert */
char *c_text; /* text of the name */
{
sym_name_entry_type *az_previous_name;
sym_name_entry_type *az_current_name;
sym_name_entry_type *az_new_name;
int l_hash_code;
int l_compare_result;
/*
** algorithm keeps 2 pointers, one for the previous name and one
** for the current name. This permits easy insertion of a new name
*/
/* obtain the hash code of for the name */
l_hash_code = hash_function( l_length, c_text );
/*
** chain along hash chain looking for symbol - exit loop under 3 condition
** 1) come to the end of the chain: insert new node on end
** 2) find symbol: return this symbol
** 3) find node > symbol: insert new node prior to current node
*/
for (az_current_name = sym_az_hash_table[ l_hash_code ],
az_previous_name = NULL;
az_current_name != NULL;
az_previous_name = az_current_name,
az_current_name = az_current_name->az_next_name_entry)
{
l_compare_result = _compare(c_text, az_current_name->c_text);
if (l_compare_result == 0) /* c_text = current name */
{
/* found the name we are looking for */
return az_current_name;
}
if (l_compare_result > 0) /* c_text > current name */
{
/* exit the loop to insert just prior to current name */
goto insert_name;
}
}
insert_name:
/*
** name is not in the table so it must be inserted between the
** az_previous_name and az_current_name entries.
*/
/* allocate and initialize the name entry */
az_new_name = (sym_name_entry_type *)
sem_allocate_node (sym_k_name_entry,
sym_k_name_entry_size + l_length + 1);
az_new_name->header.b_type = l_length; /* b_type holds length */
az_new_name->az_object = NULL;
az_new_name->az_cycle_id = 0;
_move( az_new_name->c_text, c_text, l_length+1 );
/*
** link the name entry into the hash table
*/
az_new_name->az_next_name_entry = az_current_name;
if (az_previous_name == NULL)
sym_az_hash_table[ l_hash_code ] = az_new_name;
else
az_previous_name->az_next_name_entry =
(sym_name_entry_type *) az_new_name;
return az_new_name;
}
/*
**++
** FUNCTIONAL DESCRIPTION:
**
** This procedure is a hashing function. It takes a length and a
** pointer to a value. Using this value as a string, the function
** returns an integer in the range of 0 to sym_k_hash_table_limit-1.
**
** FORMAL PARAMETERS:
**
** l_length length of the value in bytes not including null
** c_value a null terminated string
**
** IMPLICIT INPUTS:
**
** sym_k_hash_table_limit
**
** IMPLICIT OUTPUTS:
**
** none
**
** FUNCTION VALUE:
**
** integer (the hash code) in range 0 to sym_k_hash_table_limit-1
**
** SIDE EFFECTS:
**
** none
**
**--
**/
int hash_function(l_length, c_value)
int l_length;
char *c_value;
{
#ifdef WORD64
#define _shift 3
static unsigned int XmConst mask[ 8 ] =
{ 0x00000000000000FF, 0x000000000000FFFF,
0x0000000000FFFFFF, 0x00000000FFFFFFFF,
0x00000000FFFFFFFF, 0x0000FFFFFFFFFFFF,
0x00FFFFFFFFFFFFFF, 0xFFFFFFFFFFFFFFFF, };
#elif defined (LONG64)
#define _shift 3
static long XmConst mask[ 8 ] =
{ 0x00000000000000FF, 0x000000000000FFFF,
0x0000000000FFFFFF, 0x00000000FFFFFFFF,
0x00000000FFFFFFFF, 0x0000FFFFFFFFFFFF,
0x00FFFFFFFFFFFFFF, 0xFFFFFFFFFFFFFFFF, };
#else
#define _shift 2
static unsigned int XmConst mask[ 4 ] =
{ 0x000000FF, 0x0000FFFF, 0x00FFFFFF, 0xFFFFFFFF };
#endif
#ifdef LONG64
long l_hash_code;
long al_value[20];
#else
int l_hash_code;
int al_value[20];
#endif
int l_limit;
int l_extra;
int i;
l_limit = (l_length-1) >> _shift; /* divide by wordsize */
l_extra = (l_length-1) & _slm; /* remainder from divide by wordsize */
#ifdef LONG64
bzero((char *)al_value, sizeof(long) * 20);
#else
bzero((char *)al_value, sizeof(int) * 20);
#endif
strncpy((char *)al_value, c_value, l_length);
l_hash_code = 0;
for (i = 0; i < l_limit; i++)
{
l_hash_code = l_hash_code ^ al_value[ i ];
}
l_hash_code = l_hash_code ^ (al_value[ i ] & mask[ l_extra ]);
return (int)(l_hash_code % sym_k_hash_table_limit);
}
#if debug_version
/*
**++
** FUNCTIONAL DESCRIPTION:
**
** This debugging routine will dump out the name entries linked
** from the hash table.
**
** FORMAL PARAMETERS:
**
** none
**
** IMPLICIT INPUTS:
**
** sym_az_hash_table
**
** IMPLICIT OUTPUTS:
**
** none
**
** FUNCTION VALUE:
**
** void
**
** SIDE EFFECTS:
**
** prints out the hash table
**
**--
**/
void sym_dump_hash_table()
{
int i;
int total_count;
int max_length;
int empty_count;
total_count = 0;
empty_count = 0;
max_length = 0;
for (i=0; i<sym_k_hash_table_limit; i++)
{
int bucket_count;
sym_name_entry_type *az_name;
bucket_count = 0;
for (az_name = sym_az_hash_table[ i ];
az_name != NULL;
az_name = az_name->az_next_name_entry)
{
bucket_count++;
_debug_output("\t %s \n", az_name->c_text);
}
total_count += bucket_count;
if (bucket_count == 0)
empty_count++;
max_length = ( max_length > bucket_count )? max_length : bucket_count;
_debug_output("bucket: %d length: %d\n", i, bucket_count);
}
_debug_output("name count: %d \n empty count: %d \n max length: %d",
total_count, empty_count, max_length );
}
#endif