* [PATCH] lib/boilerplate: Allow building with GCC 10.2
@ 2020-11-11 14:59 florian.bezdeka
2020-11-12 6:58 ` Jan Kiszka
0 siblings, 1 reply; 5+ messages in thread
From: florian.bezdeka @ 2020-11-11 14:59 UTC (permalink / raw)
To: xenomai
When trying to compile using GCC 10.2 the following warning / error
appeared:
iniparser/iniparser.c: In function ‘iniparser_load’:
iniparser/iniparser.c:616:13: error: ‘sprintf’ arguments 3, 4 may overlap destination object ‘buf’ [-Werror=restrict]
616 | sprintf(tmp, "%s:%s", section, key);
| ^~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
According to POSIX and C99 standard using sprintf() on overlapping
buffers may lead to undefined behavior, so using a dedicated buffer for
the destination (first argument to sprintf) is necessary.
While at it:
- Fixed some indentations
- Fixed some error handling code paths
Some resources were not properly cleaned up in error conditions
- Fixed buffer initializations: The last bytes were never set to zero
Fixes: 67803dce543e ("lib/boilerplate: add iniparser to boilerplate")
Signed-off-by: Florian Bezdeka <florian.bezdeka@siemens.com>
---
lib/boilerplate/iniparser/iniparser.c | 39 ++++++++++++++++++---------
1 file changed, 27 insertions(+), 12 deletions(-)
diff --git a/lib/boilerplate/iniparser/iniparser.c b/lib/boilerplate/iniparser/iniparser.c
index 5b2094a00..b79f965a6 100644
--- a/lib/boilerplate/iniparser/iniparser.c
+++ b/lib/boilerplate/iniparser/iniparser.c
@@ -551,25 +551,37 @@ dictionary * iniparser_load(const char * ininame)
dict = dictionary_new(0) ;
if (!dict) {
fclose(in);
- errno = ENOMEM;
+ errno = ENOMEM;
return NULL ;
}
- buf = malloc((ASCIILINESZ+1) * 5);
+ buf = malloc((ASCIILINESZ+1) * 4);
if (buf == NULL) {
- errno = -ENOMEM;
- return NULL;
+ dictionary_del(dict);
+ fclose(in);
+ errno = ENOMEM;
+ return NULL;
}
+
+ tmp = malloc(ASCIILINESZ + 1);
+ if (!tmp) {
+ free(buf);
+ dictionary_del(dict);
+ fclose(in);
+ errno = ENOMEM;
+ return NULL;
+ }
+
line = buf;
section = line + ASCIILINESZ + 1;
key = section + ASCIILINESZ + 1;
- tmp = key + ASCIILINESZ + 1;
- val = tmp + ASCIILINESZ + 1;
+ val = key + ASCIILINESZ + 1;
- memset(line, 0, ASCIILINESZ);
- memset(section, 0, ASCIILINESZ);
- memset(key, 0, ASCIILINESZ);
- memset(val, 0, ASCIILINESZ);
+ memset(line, 0, ASCIILINESZ + 1);
+ memset(section, 0, ASCIILINESZ + 1);
+ memset(key, 0, ASCIILINESZ + 1);
+ memset(val, 0, ASCIILINESZ + 1);
+ memset(tmp, 0, ASCIILINESZ + 1);
last=0 ;
while (fgets(line+last, ASCIILINESZ-last, in)!=NULL) {
@@ -585,8 +597,9 @@ dictionary * iniparser_load(const char * ininame)
#endif
dictionary_del(dict);
fclose(in);
- free(buf);
- errno = EINVAL;
+ free(buf);
+ free(tmp);
+ errno = EINVAL;
return NULL ;
}
/* Get rid of \n and spaces at end of line */
@@ -613,6 +626,7 @@ dictionary * iniparser_load(const char * ininame)
break ;
case LINE_VALUE:
+ /* Already checked for overflows, see above */
sprintf(tmp, "%s:%s", section, key);
errs = dictionary_set(dict, tmp, val) ;
break ;
@@ -642,6 +656,7 @@ dictionary * iniparser_load(const char * ininame)
}
fclose(in);
free(buf);
+ free(tmp);
if (errs) {
dictionary_del(dict);
dict = NULL ;
--
2.26.2
^ permalink raw reply related [flat|nested] 5+ messages in thread
* Re: [PATCH] lib/boilerplate: Allow building with GCC 10.2
2020-11-11 14:59 [PATCH] lib/boilerplate: Allow building with GCC 10.2 florian.bezdeka
@ 2020-11-12 6:58 ` Jan Kiszka
2020-11-12 8:49 ` florian.bezdeka
0 siblings, 1 reply; 5+ messages in thread
From: Jan Kiszka @ 2020-11-12 6:58 UTC (permalink / raw)
To: florian.bezdeka, xenomai
Interesting corner of our code base...
On 11.11.20 15:59, florian.bezdeka--- via Xenomai wrote:
> When trying to compile using GCC 10.2 the following warning / error
> appeared:
>
> iniparser/iniparser.c: In function ‘iniparser_load’:
> iniparser/iniparser.c:616:13: error: ‘sprintf’ arguments 3, 4 may overlap destination object ‘buf’ [-Werror=restrict]
> 616 | sprintf(tmp, "%s:%s", section, key);
> | ^~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
>
Strange, cannot reproduce this report here with
gcc-10 (SUSE Linux) 10.2.1 20200825 [revision
c0746a1beb1ba073c7981eb09f55b3d993b32e5c]
even when explicitly adding -Wrestrict.
> According to POSIX and C99 standard using sprintf() on overlapping
> buffers may lead to undefined behavior, so using a dedicated buffer for
> the destination (first argument to sprintf) is necessary.
>
> While at it:
> - Fixed some indentations
> - Fixed some error handling code paths
> Some resources were not properly cleaned up in error conditions
> - Fixed buffer initializations: The last bytes were never set to zero
>
We likely first of all want a resync with
https://github.com/ndevilla/iniparser. If there is an upstream issue,
maybe fix there first.
> Fixes: 67803dce543e ("lib/boilerplate: add iniparser to boilerplate")
> Signed-off-by: Florian Bezdeka <florian.bezdeka@siemens.com>
> ---
> lib/boilerplate/iniparser/iniparser.c | 39 ++++++++++++++++++---------
> 1 file changed, 27 insertions(+), 12 deletions(-)
>
> diff --git a/lib/boilerplate/iniparser/iniparser.c b/lib/boilerplate/iniparser/iniparser.c
> index 5b2094a00..b79f965a6 100644
> --- a/lib/boilerplate/iniparser/iniparser.c
> +++ b/lib/boilerplate/iniparser/iniparser.c
> @@ -551,25 +551,37 @@ dictionary * iniparser_load(const char * ininame)
> dict = dictionary_new(0) ;
> if (!dict) {
> fclose(in);
> - errno = ENOMEM;
> + errno = ENOMEM;
> return NULL ;
> }
>
> - buf = malloc((ASCIILINESZ+1) * 5);
> + buf = malloc((ASCIILINESZ+1) * 4);
> if (buf == NULL) {
> - errno = -ENOMEM;
> - return NULL;
> + dictionary_del(dict);
> + fclose(in);
> + errno = ENOMEM;
> + return NULL;
> }
> +
> + tmp = malloc(ASCIILINESZ + 1);
> + if (!tmp) {
> + free(buf);
> + dictionary_del(dict);
> + fclose(in);
> + errno = ENOMEM;
> + return NULL;
> + }
> +
This split-up is just silencing the compiler, right? It does no take
away the possibility that we overrun the target buffer in sprintf
(section+key > ASCIILINESZ). Upstream seems to have addressed that.
But it also does not use snprintf yet. Likely with the argument that
buffer sizes are under full control.
> line = buf;
> section = line + ASCIILINESZ + 1;
> key = section + ASCIILINESZ + 1;
> - tmp = key + ASCIILINESZ + 1;
> - val = tmp + ASCIILINESZ + 1;
> + val = key + ASCIILINESZ + 1;
>
> - memset(line, 0, ASCIILINESZ);
> - memset(section, 0, ASCIILINESZ);
> - memset(key, 0, ASCIILINESZ);
> - memset(val, 0, ASCIILINESZ);
> + memset(line, 0, ASCIILINESZ + 1);
> + memset(section, 0, ASCIILINESZ + 1);
> + memset(key, 0, ASCIILINESZ + 1);
> + memset(val, 0, ASCIILINESZ + 1);
> + memset(tmp, 0, ASCIILINESZ + 1);
> last=0 ;
>
> while (fgets(line+last, ASCIILINESZ-last, in)!=NULL) {
> @@ -585,8 +597,9 @@ dictionary * iniparser_load(const char * ininame)
> #endif
> dictionary_del(dict);
> fclose(in);
> - free(buf);
> - errno = EINVAL;
> + free(buf);
> + free(tmp);
> + errno = EINVAL;
> return NULL ;
> }
> /* Get rid of \n and spaces at end of line */
> @@ -613,6 +626,7 @@ dictionary * iniparser_load(const char * ininame)
> break ;
>
> case LINE_VALUE:
> + /* Already checked for overflows, see above */
> sprintf(tmp, "%s:%s", section, key);
> errs = dictionary_set(dict, tmp, val) ;
> break ;
> @@ -642,6 +656,7 @@ dictionary * iniparser_load(const char * ininame)
> }
> fclose(in);
> free(buf);
> + free(tmp);
> if (errs) {
> dictionary_del(dict);
> dict = NULL ;
>
Jan
--
Siemens AG, T RDA IOT
Corporate Competence Center Embedded Linux
^ permalink raw reply [flat|nested] 5+ messages in thread
* Re: [PATCH] lib/boilerplate: Allow building with GCC 10.2
2020-11-12 6:58 ` Jan Kiszka
@ 2020-11-12 8:49 ` florian.bezdeka
2020-11-12 11:45 ` [PATCH v2] lib/boilerplate/iniparser: Allow building with GCC 10.2 2020101 florian.bezdeka
0 siblings, 1 reply; 5+ messages in thread
From: florian.bezdeka @ 2020-11-12 8:49 UTC (permalink / raw)
To: xenomai, jan.kiszka
On Thu, 2020-11-12 at 07:58 +0100, Jan Kiszka wrote:
> Interesting corner of our code base...
>
> On 11.11.20 15:59, florian.bezdeka--- via Xenomai wrote:
> > When trying to compile using GCC 10.2 the following warning / error
> > appeared:
> >
> > iniparser/iniparser.c: In function ‘iniparser_load’:
> > iniparser/iniparser.c:616:13: error: ‘sprintf’ arguments 3, 4 may
> > overlap destination object ‘buf’ [-Werror=restrict]
> > 616 | sprintf(tmp, "%s:%s", section, key);
> > | ^~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
> >
>
> Strange, cannot reproduce this report here with
>
> gcc-10 (SUSE Linux) 10.2.1 20200825 [revision
> c0746a1beb1ba073c7981eb09f55b3d993b32e5c]
Here:
gcc version 10.2.1 20201016 (Red Hat 10.2.1-6) (GCC)
>
> even when explicitly adding -Wrestrict.
>
> > According to POSIX and C99 standard using sprintf() on overlapping
> > buffers may lead to undefined behavior, so using a dedicated buffer
> > for
> > the destination (first argument to sprintf) is necessary.
> >
> > While at it:
> > - Fixed some indentations
> > - Fixed some error handling code paths
> > Some resources were not properly cleaned up in error conditions
> > - Fixed buffer initializations: The last bytes were never set to
> > zero
> >
>
> We likely first of all want a resync with
> https://github.com/ndevilla/iniparser. If there is an upstream issue,
> maybe fix there first.
Oh, sorry... I did not recognize that we have an upstream involved. I
will retry with latest upstream version.
>
> > Fixes: 67803dce543e ("lib/boilerplate: add iniparser to
> > boilerplate")
> > Signed-off-by: Florian Bezdeka <florian.bezdeka@siemens.com>
> > ---
> > lib/boilerplate/iniparser/iniparser.c | 39 ++++++++++++++++++-----
> > ----
> > 1 file changed, 27 insertions(+), 12 deletions(-)
> >
> > diff --git a/lib/boilerplate/iniparser/iniparser.c
> > b/lib/boilerplate/iniparser/iniparser.c
> > index 5b2094a00..b79f965a6 100644
> > --- a/lib/boilerplate/iniparser/iniparser.c
> > +++ b/lib/boilerplate/iniparser/iniparser.c
> > @@ -551,25 +551,37 @@ dictionary * iniparser_load(const char *
> > ininame)
> > dict = dictionary_new(0) ;
> > if (!dict) {
> > fclose(in);
> > - errno = ENOMEM;
> > + errno = ENOMEM;
> > return NULL ;
> > }
> >
> > - buf = malloc((ASCIILINESZ+1) * 5);
> > + buf = malloc((ASCIILINESZ+1) * 4);
> > if (buf == NULL) {
> > - errno = -ENOMEM;
> > - return NULL;
> > + dictionary_del(dict);
> > + fclose(in);
> > + errno = ENOMEM;
> > + return NULL;
> > }
> > +
> > + tmp = malloc(ASCIILINESZ + 1);
> > + if (!tmp) {
> > + free(buf);
> > + dictionary_del(dict);
> > + fclose(in);
> > + errno = ENOMEM;
> > + return NULL;
> > + }
> > +
>
> This split-up is just silencing the compiler, right? It does no take
> away the possibility that we overrun the target buffer in sprintf
> (section+key > ASCIILINESZ). Upstream seems to have addressed that.
Silencing the compiler but overflow is handled (so taken care of) in
iniparser_line. Upstream switched to buffers on the stack in the
meantime, so error handling is simplified a bit.
>
> But it also does not use snprintf yet. Likely with the argument that
> buffer sizes are under full control.
>
> > line = buf;
> > section = line + ASCIILINESZ + 1;
> > key = section + ASCIILINESZ + 1;
> > - tmp = key + ASCIILINESZ + 1;
> > - val = tmp + ASCIILINESZ + 1;
> > + val = key + ASCIILINESZ + 1;
> >
> > - memset(line, 0, ASCIILINESZ);
> > - memset(section, 0, ASCIILINESZ);
> > - memset(key, 0, ASCIILINESZ);
> > - memset(val, 0, ASCIILINESZ);
> > + memset(line, 0, ASCIILINESZ + 1);
> > + memset(section, 0, ASCIILINESZ + 1);
> > + memset(key, 0, ASCIILINESZ + 1);
> > + memset(val, 0, ASCIILINESZ + 1);
> > + memset(tmp, 0, ASCIILINESZ + 1);
> > last=0 ;
> >
> > while (fgets(line+last, ASCIILINESZ-last, in)!=NULL) {
> > @@ -585,8 +597,9 @@ dictionary * iniparser_load(const char *
> > ininame)
> > #endif
> > dictionary_del(dict);
> > fclose(in);
> > - free(buf);
> > - errno = EINVAL;
> > + free(buf);
> > + free(tmp);
> > + errno = EINVAL;
> > return NULL ;
> > }
> > /* Get rid of \n and spaces at end of line */
> > @@ -613,6 +626,7 @@ dictionary * iniparser_load(const char *
> > ininame)
> > break ;
> >
> > case LINE_VALUE:
> > + /* Already checked for overflows, see above */
> > sprintf(tmp, "%s:%s", section, key);
> > errs = dictionary_set(dict, tmp, val) ;
> > break ;
> > @@ -642,6 +656,7 @@ dictionary * iniparser_load(const char *
> > ininame)
> > }
> > fclose(in);
> > free(buf);
> > + free(tmp);
> > if (errs) {
> > dictionary_del(dict);
> > dict = NULL ;
> >
>
> Jan
^ permalink raw reply [flat|nested] 5+ messages in thread
* [PATCH v2] lib/boilerplate/iniparser: Allow building with GCC 10.2 2020101
2020-11-12 8:49 ` florian.bezdeka
@ 2020-11-12 11:45 ` florian.bezdeka
2020-11-12 16:19 ` Jan Kiszka
0 siblings, 1 reply; 5+ messages in thread
From: florian.bezdeka @ 2020-11-12 11:45 UTC (permalink / raw)
To: xenomai, jan.kiszka
From: Florian Bezdeka <florian.bezdeka@siemens.com>
Updating to upstream revision f858275f7f307eecba84c2f5429483f9f28007f8.
The reason for updating was the following compiler error when trying
to compile with GCC 10.2 10.2.1 20201016. As it turned out the problem
was already addressed upstream:
iniparser/iniparser.c: In function ‘iniparser_load’:
iniparser/iniparser.c:616:13: error: ‘sprintf’ arguments 3, 4 may
overlap destination object ‘buf’ [-Werror=restrict]
616 | sprintf(tmp, "%s:%s", section, key);
| ^~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
I reviewed especially the API changes. Most of them are cleanups only
but two things should be pointed out:
- The type of the size field of struct _dictionary_ changed from int
to ssize_t. The only user of this struct is
lib/analogy/calibration.c which uses this structure for internal
things only. It is never exposed to any public API so updating is
OK and fully backward compatible.
- dictionary_new changed its signature
from dictionary_new(int size)
to dictionary_new(size_t size).
This function is not part of any public API. So updating does not
break backward compatibility.
Signed-off-by: Florian Bezdeka <florian.bezdeka@siemens.com>
---
lib/boilerplate/iniparser/dictionary.c | 409 ++++++++++----------
lib/boilerplate/iniparser/dictionary.h | 43 ++-
lib/boilerplate/iniparser/iniparser.c | 491 +++++++++++++++++--------
lib/boilerplate/iniparser/iniparser.h | 131 +++++--
4 files changed, 646 insertions(+), 428 deletions(-)
diff --git a/lib/boilerplate/iniparser/dictionary.c b/lib/boilerplate/iniparser/dictionary.c
index 5299b77ed..cb7ccd49e 100644
--- a/lib/boilerplate/iniparser/dictionary.c
+++ b/lib/boilerplate/iniparser/dictionary.c
@@ -1,10 +1,8 @@
/*-------------------------------------------------------------------------*/
/**
- @file dictionary.c
- @author N. Devillard
- @date Sep 2007
- @version $Revision: 1.27 $
- @brief Implements a dictionary for string variables.
+ @file dictionary.c
+ @author N. Devillard
+ @brief Implements a dictionary for string variables.
This module implements a simple dictionary object, i.e. a list
of string/string associations. This object is useful to store e.g.
@@ -12,12 +10,8 @@
*/
/*--------------------------------------------------------------------------*/
-/*
- $Id: dictionary.c,v 1.27 2007-11-23 21:39:18 ndevilla Exp $
- $Revision: 1.27 $
-*/
/*---------------------------------------------------------------------------
- Includes
+ Includes
---------------------------------------------------------------------------*/
#include "dictionary.h"
@@ -27,33 +21,18 @@
#include <unistd.h>
/** Maximum value size for integers and doubles. */
-#define MAXVALSZ 1024
+#define MAXVALSZ 1024
/** Minimal allocated number of entries in a dictionary */
-#define DICTMINSZ 128
+#define DICTMINSZ 128
/** Invalid key token */
#define DICT_INVALID_KEY ((char*)-1)
/*---------------------------------------------------------------------------
- Private functions
+ Private functions
---------------------------------------------------------------------------*/
-/* Doubles the allocated size associated to a pointer */
-/* 'size' is the current allocated size. */
-static void * mem_double(void * ptr, int size)
-{
- void * newptr ;
-
- newptr = calloc(2*size, 1);
- if (newptr==NULL) {
- return NULL ;
- }
- memcpy(newptr, ptr, size);
- free(ptr);
- return newptr ;
-}
-
/*-------------------------------------------------------------------------*/
/**
@brief Duplicate a string
@@ -67,23 +46,68 @@ static void * mem_double(void * ptr, int size)
static char * xstrdup(const char * s)
{
char * t ;
+ size_t len ;
if (!s)
return NULL ;
- t = malloc(strlen(s)+1) ;
+
+ len = strlen(s) + 1 ;
+ t = (char*) malloc(len) ;
if (t) {
- strcpy(t,s);
+ memcpy(t, s, len) ;
}
return t ;
}
+/*-------------------------------------------------------------------------*/
+/**
+ @brief Double the size of the dictionary
+ @param d Dictionary to grow
+ @return This function returns non-zero in case of failure
+ */
+/*--------------------------------------------------------------------------*/
+static int dictionary_grow(dictionary * d)
+{
+ char ** new_val ;
+ char ** new_key ;
+ unsigned * new_hash ;
+
+ new_val = (char**) calloc(d->size * 2, sizeof *d->val);
+ new_key = (char**) calloc(d->size * 2, sizeof *d->key);
+ new_hash = (unsigned*) calloc(d->size * 2, sizeof *d->hash);
+ if (!new_val || !new_key || !new_hash) {
+ /* An allocation failed, leave the dictionary unchanged */
+ if (new_val)
+ free(new_val);
+ if (new_key)
+ free(new_key);
+ if (new_hash)
+ free(new_hash);
+ return -1 ;
+ }
+ /* Initialize the newly allocated space */
+ memcpy(new_val, d->val, d->size * sizeof(char *));
+ memcpy(new_key, d->key, d->size * sizeof(char *));
+ memcpy(new_hash, d->hash, d->size * sizeof(unsigned));
+ /* Delete previous data */
+ free(d->val);
+ free(d->key);
+ free(d->hash);
+ /* Actually update the dictionary */
+ d->size *= 2 ;
+ d->val = new_val;
+ d->key = new_key;
+ d->hash = new_hash;
+ return 0 ;
+}
+
/*---------------------------------------------------------------------------
- Function codes
+ Function codes
---------------------------------------------------------------------------*/
/*-------------------------------------------------------------------------*/
/**
- @brief Compute the hash key for a string.
- @param key Character string to use for key.
- @return 1 unsigned int on at least 32 bits.
+ @brief Compute the hash key for a string.
+ @param key Character string to use for key.
+ @return 1 unsigned int on at least 32 bits.
This hash function has been taken from an Article in Dr Dobbs Journal.
This is normally a collision-free function, distributing keys evenly.
@@ -93,84 +117,88 @@ static char * xstrdup(const char * s)
/*--------------------------------------------------------------------------*/
unsigned dictionary_hash(const char * key)
{
- int len ;
- unsigned hash ;
- int i ;
-
- len = strlen(key);
- for (hash=0, i=0 ; i<len ; i++) {
- hash += (unsigned)key[i] ;
- hash += (hash<<10);
- hash ^= (hash>>6) ;
- }
- hash += (hash <<3);
- hash ^= (hash >>11);
- hash += (hash <<15);
- return hash ;
+ size_t len ;
+ unsigned hash ;
+ size_t i ;
+
+ if (!key)
+ return 0 ;
+
+ len = strlen(key);
+ for (hash=0, i=0 ; i<len ; i++) {
+ hash += (unsigned)key[i] ;
+ hash += (hash<<10);
+ hash ^= (hash>>6) ;
+ }
+ hash += (hash <<3);
+ hash ^= (hash >>11);
+ hash += (hash <<15);
+ return hash ;
}
/*-------------------------------------------------------------------------*/
/**
- @brief Create a new dictionary object.
- @param size Optional initial size of the dictionary.
- @return 1 newly allocated dictionary objet.
+ @brief Create a new dictionary object.
+ @param size Optional initial size of the dictionary.
+ @return 1 newly allocated dictionary objet.
This function allocates a new dictionary object of given size and returns
it. If you do not know in advance (roughly) the number of entries in the
dictionary, give size=0.
*/
-/*--------------------------------------------------------------------------*/
-dictionary * dictionary_new(int size)
+/*-------------------------------------------------------------------------*/
+dictionary * dictionary_new(size_t size)
{
- dictionary * d ;
-
- /* If no size was specified, allocate space for DICTMINSZ */
- if (size<DICTMINSZ) size=DICTMINSZ ;
-
- if (!(d = (dictionary *)calloc(1, sizeof(dictionary)))) {
- return NULL;
- }
- d->size = size ;
- d->val = (char **)calloc(size, sizeof(char*));
- d->key = (char **)calloc(size, sizeof(char*));
- d->hash = (unsigned int *)calloc(size, sizeof(unsigned));
- return d ;
+ dictionary * d ;
+
+ /* If no size was specified, allocate space for DICTMINSZ */
+ if (size<DICTMINSZ) size=DICTMINSZ ;
+
+ d = (dictionary*) calloc(1, sizeof *d) ;
+
+ if (d) {
+ d->size = size ;
+ d->val = (char**) calloc(size, sizeof *d->val);
+ d->key = (char**) calloc(size, sizeof *d->key);
+ d->hash = (unsigned*) calloc(size, sizeof *d->hash);
+ }
+ return d ;
}
/*-------------------------------------------------------------------------*/
/**
- @brief Delete a dictionary object
- @param d dictionary object to deallocate.
- @return void
+ @brief Delete a dictionary object
+ @param d dictionary object to deallocate.
+ @return void
Deallocate a dictionary object and all memory associated to it.
*/
/*--------------------------------------------------------------------------*/
void dictionary_del(dictionary * d)
{
- int i ;
-
- if (d==NULL) return ;
- for (i=0 ; i<d->size ; i++) {
- if (d->key[i]!=NULL)
- free(d->key[i]);
- if (d->val[i]!=NULL)
- free(d->val[i]);
- }
- free(d->val);
- free(d->key);
- free(d->hash);
- free(d);
- return ;
+ ssize_t i ;
+
+ if (d==NULL) return ;
+ for (i=0 ; i<d->size ; i++) {
+ if (d->key[i]!=NULL)
+ free(d->key[i]);
+ if (d->val[i]!=NULL)
+ free(d->val[i]);
+ }
+ free(d->val);
+ free(d->key);
+ free(d->hash);
+ free(d);
+ return ;
}
/*-------------------------------------------------------------------------*/
/**
- @brief Get a value from a dictionary.
- @param d dictionary object to search.
- @param key Key to look for in the dictionary.
+ @brief Get a value from a dictionary.
+ @param d dictionary object to search.
+ @param key Key to look for in the dictionary.
@param def Default value to return if key not found.
- @return 1 pointer to internally allocated character string.
+ @return 1 pointer to internally allocated character string.
This function locates a key in a dictionary and returns a pointer to its
value, or the passed 'def' pointer if no such key can be found in
@@ -178,24 +206,24 @@ void dictionary_del(dictionary * d)
dictionary object, you should not try to free it or modify it.
*/
/*--------------------------------------------------------------------------*/
-const char * dictionary_get(dictionary * d, const char * key, const char * def)
+const char * dictionary_get(const dictionary * d, const char * key, const char * def)
{
- unsigned hash ;
- int i ;
+ unsigned hash ;
+ ssize_t i ;
- hash = dictionary_hash(key);
- for (i=0 ; i<d->size ; i++) {
+ hash = dictionary_hash(key);
+ for (i=0 ; i<d->size ; i++) {
if (d->key[i]==NULL)
continue ;
/* Compare hash */
- if (hash==d->hash[i]) {
+ if (hash==d->hash[i]) {
/* Compare string, to avoid hash collisions */
if (!strcmp(key, d->key[i])) {
- return d->val[i] ;
- }
- }
- }
- return def ;
+ return d->val[i] ;
+ }
+ }
+ }
+ return def ;
}
/*-------------------------------------------------------------------------*/
@@ -226,66 +254,57 @@ const char * dictionary_get(dictionary * d, const char * key, const char * def)
/*--------------------------------------------------------------------------*/
int dictionary_set(dictionary * d, const char * key, const char * val)
{
- int i ;
- unsigned hash ;
-
- if (d==NULL || key==NULL) return -1 ;
-
- /* Compute hash for this key */
- hash = dictionary_hash(key) ;
- /* Find if value is already in dictionary */
- if (d->n>0) {
- for (i=0 ; i<d->size ; i++) {
+ ssize_t i ;
+ unsigned hash ;
+
+ if (d==NULL || key==NULL) return -1 ;
+
+ /* Compute hash for this key */
+ hash = dictionary_hash(key) ;
+ /* Find if value is already in dictionary */
+ if (d->n>0) {
+ for (i=0 ; i<d->size ; i++) {
if (d->key[i]==NULL)
continue ;
- if (hash==d->hash[i]) { /* Same hash value */
- if (!strcmp(key, d->key[i])) { /* Same key */
- /* Found a value: modify and return */
- if (d->val[i]!=NULL)
- free(d->val[i]);
- d->val[i] = val ? xstrdup(val) : NULL ;
+ if (hash==d->hash[i]) { /* Same hash value */
+ if (!strcmp(key, d->key[i])) { /* Same key */
+ /* Found a value: modify and return */
+ if (d->val[i]!=NULL)
+ free(d->val[i]);
+ d->val[i] = (val ? xstrdup(val) : NULL);
/* Value has been modified: return */
- return 0 ;
- }
- }
- }
- }
- /* Add a new value */
- /* See if dictionary needs to grow */
- if (d->n==d->size) {
-
- /* Reached maximum size: reallocate dictionary */
- d->val = (char **)mem_double(d->val, d->size * sizeof(char*)) ;
- d->key = (char **)mem_double(d->key, d->size * sizeof(char*)) ;
- d->hash = (unsigned int *)mem_double(d->hash, d->size * sizeof(unsigned)) ;
- if ((d->val==NULL) || (d->key==NULL) || (d->hash==NULL)) {
- /* Cannot grow dictionary */
- return -1 ;
+ return 0 ;
+ }
+ }
}
- /* Double size */
- d->size *= 2 ;
- }
+ }
+ /* Add a new value */
+ /* See if dictionary needs to grow */
+ if (d->n==d->size) {
+ /* Reached maximum size: reallocate dictionary */
+ if (dictionary_grow(d) != 0)
+ return -1;
+ }
- /* Insert key in the first empty slot */
- for (i=0 ; i<d->size ; i++) {
- if (d->key[i]==NULL) {
- /* Add key here */
- break ;
- }
+ /* Insert key in the first empty slot. Start at d->n and wrap at
+ d->size. Because d->n < d->size this will necessarily
+ terminate. */
+ for (i=d->n ; d->key[i] ; ) {
+ if(++i == d->size) i = 0;
}
- /* Copy key */
- d->key[i] = xstrdup(key);
- d->val[i] = val ? xstrdup(val) : NULL ;
- d->hash[i] = hash;
- d->n ++ ;
- return 0 ;
+ /* Copy key */
+ d->key[i] = xstrdup(key);
+ d->val[i] = (val ? xstrdup(val) : NULL) ;
+ d->hash[i] = hash;
+ d->n ++ ;
+ return 0 ;
}
/*-------------------------------------------------------------------------*/
/**
- @brief Delete a key in a dictionary
- @param d dictionary object to modify.
- @param key Key to remove.
+ @brief Delete a key in a dictionary
+ @param d dictionary object to modify.
+ @param key Key to remove.
@return void
This function deletes a key in a dictionary. Nothing is done if the
@@ -294,26 +313,26 @@ int dictionary_set(dictionary * d, const char * key, const char * val)
/*--------------------------------------------------------------------------*/
void dictionary_unset(dictionary * d, const char * key)
{
- unsigned hash ;
- int i ;
+ unsigned hash ;
+ ssize_t i ;
- if (key == NULL) {
- return;
- }
+ if (key == NULL || d == NULL) {
+ return;
+ }
- hash = dictionary_hash(key);
- for (i=0 ; i<d->size ; i++) {
+ hash = dictionary_hash(key);
+ for (i=0 ; i<d->size ; i++) {
if (d->key[i]==NULL)
continue ;
/* Compare hash */
- if (hash==d->hash[i]) {
+ if (hash==d->hash[i]) {
/* Compare string, to avoid hash collisions */
if (!strcmp(key, d->key[i])) {
/* Found key */
break ;
- }
- }
- }
+ }
+ }
+ }
if (i>=d->size)
/* Key not found */
return ;
@@ -331,75 +350,31 @@ void dictionary_unset(dictionary * d, const char * key)
/*-------------------------------------------------------------------------*/
/**
- @brief Dump a dictionary to an opened file pointer.
- @param d Dictionary to dump
- @param out Opened file pointer.
- @return void
+ @brief Dump a dictionary to an opened file pointer.
+ @param d Dictionary to dump
+ @param f Opened file pointer.
+ @return void
Dumps a dictionary onto an opened file pointer. Key pairs are printed out
as @c [Key]=[Value], one per line. It is Ok to provide stdout or stderr as
output file pointers.
*/
/*--------------------------------------------------------------------------*/
-void dictionary_dump(dictionary * d, FILE * out)
+void dictionary_dump(const dictionary * d, FILE * out)
{
- int i ;
-
- if (d==NULL || out==NULL) return ;
- if (d->n<1) {
- fprintf(out, "empty dictionary\n");
- return ;
- }
- for (i=0 ; i<d->size ; i++) {
+ ssize_t i ;
+
+ if (d==NULL || out==NULL) return ;
+ if (d->n<1) {
+ fprintf(out, "empty dictionary\n");
+ return ;
+ }
+ for (i=0 ; i<d->size ; i++) {
if (d->key[i]) {
fprintf(out, "%20s\t[%s]\n",
d->key[i],
d->val[i] ? d->val[i] : "UNDEF");
}
- }
- return ;
-}
-
-
-/* Test code */
-#ifdef TESTDIC
-#define NVALS 20000
-int main(int argc, char *argv[])
-{
- dictionary * d ;
- char * val ;
- int i ;
- char cval[90] ;
-
- /* Allocate dictionary */
- printf("allocating...\n");
- d = dictionary_new(0);
-
- /* Set values in dictionary */
- printf("setting %d values...\n", NVALS);
- for (i=0 ; i<NVALS ; i++) {
- sprintf(cval, "%04d", i);
- dictionary_set(d, cval, "salut");
- }
- printf("getting %d values...\n", NVALS);
- for (i=0 ; i<NVALS ; i++) {
- sprintf(cval, "%04d", i);
- val = dictionary_get(d, cval, DICT_INVALID_KEY);
- if (val==DICT_INVALID_KEY) {
- printf("cannot get value for key [%s]\n", cval);
- }
- }
- printf("unsetting %d values...\n", NVALS);
- for (i=0 ; i<NVALS ; i++) {
- sprintf(cval, "%04d", i);
- dictionary_unset(d, cval);
- }
- if (d->n != 0) {
- printf("error deleting values\n");
}
- printf("deallocating...\n");
- dictionary_del(d);
- return 0 ;
+ return ;
}
-#endif
-/* vim: set ts=4 et sw=4 tw=75 */
diff --git a/lib/boilerplate/iniparser/dictionary.h b/lib/boilerplate/iniparser/dictionary.h
index fa4dcb727..d04b6ce71 100644
--- a/lib/boilerplate/iniparser/dictionary.h
+++ b/lib/boilerplate/iniparser/dictionary.h
@@ -3,8 +3,6 @@
/**
@file dictionary.h
@author N. Devillard
- @date Sep 2007
- @version $Revision: 1.12 $
@brief Implements a dictionary for string variables.
This module implements a simple dictionary object, i.e. a list
@@ -13,18 +11,11 @@
*/
/*--------------------------------------------------------------------------*/
-/*
- $Id: dictionary.h,v 1.12 2007-11-23 21:37:00 ndevilla Exp $
- $Author: ndevilla $
- $Date: 2007-11-23 21:37:00 $
- $Revision: 1.12 $
-*/
-
#ifndef _DICTIONARY_H_
#define _DICTIONARY_H_
/*---------------------------------------------------------------------------
- Includes
+ Includes
---------------------------------------------------------------------------*/
#include <stdio.h>
@@ -32,14 +23,18 @@
#include <string.h>
#include <unistd.h>
+#ifdef __cplusplus
+extern "C" {
+#endif
+
/*---------------------------------------------------------------------------
- New types
+ New types
---------------------------------------------------------------------------*/
/*-------------------------------------------------------------------------*/
/**
- @brief Dictionary object
+ @brief Dictionary object
This object contains a list of string/string associations. Each
association is identified by a unique string key. Looking up values
@@ -48,16 +43,16 @@
*/
/*-------------------------------------------------------------------------*/
typedef struct _dictionary_ {
- int n ; /** Number of entries in dictionary */
- int size ; /** Storage size */
- char ** val ; /** List of string values */
- char ** key ; /** List of string keys */
- unsigned * hash ; /** List of hash values for keys */
+ int n ; /** Number of entries in dictionary */
+ ssize_t size ; /** Storage size */
+ char ** val ; /** List of string values */
+ char ** key ; /** List of string keys */
+ unsigned * hash ; /** List of hash values for keys */
} dictionary ;
/*---------------------------------------------------------------------------
- Function prototypes
+ Function prototypes
---------------------------------------------------------------------------*/
/*-------------------------------------------------------------------------*/
@@ -85,7 +80,7 @@ unsigned dictionary_hash(const char * key);
dictionary, give size=0.
*/
/*--------------------------------------------------------------------------*/
-dictionary * dictionary_new(int size);
+dictionary * dictionary_new(size_t size);
/*-------------------------------------------------------------------------*/
/**
@@ -112,7 +107,7 @@ void dictionary_del(dictionary * vd);
dictionary object, you should not try to free it or modify it.
*/
/*--------------------------------------------------------------------------*/
-const char * dictionary_get(dictionary * d, const char * key, const char * def);
+const char * dictionary_get(const dictionary * d, const char * key, const char * def);
/*-------------------------------------------------------------------------*/
@@ -161,7 +156,7 @@ void dictionary_unset(dictionary * d, const char * key);
/**
@brief Dump a dictionary to an opened file pointer.
@param d Dictionary to dump
- @param out Opened file pointer.
+ @param f Opened file pointer.
@return void
Dumps a dictionary onto an opened file pointer. Key pairs are printed out
@@ -169,6 +164,10 @@ void dictionary_unset(dictionary * d, const char * key);
output file pointers.
*/
/*--------------------------------------------------------------------------*/
-void dictionary_dump(dictionary * d, FILE * out);
+void dictionary_dump(const dictionary * d, FILE * out);
+
+#ifdef __cplusplus
+}
+#endif
#endif
diff --git a/lib/boilerplate/iniparser/iniparser.c b/lib/boilerplate/iniparser/iniparser.c
index 5b2094a00..f1d165896 100644
--- a/lib/boilerplate/iniparser/iniparser.c
+++ b/lib/boilerplate/iniparser/iniparser.c
@@ -3,19 +3,12 @@
/**
@file iniparser.c
@author N. Devillard
- @date Sep 2007
- @version 3.0
@brief Parser for ini files.
*/
/*--------------------------------------------------------------------------*/
-/*
- $Id: iniparser.c,v 2.18 2008-01-03 18:35:39 ndevilla Exp $
- $Revision: 2.18 $
- $Date: 2008-01-03 18:35:39 $
-*/
/*---------------------------- Includes ------------------------------------*/
#include <ctype.h>
-#include <errno.h>
+#include <stdarg.h>
#include "iniparser.h"
/*---------------------------- Defines -------------------------------------*/
@@ -39,65 +32,115 @@ typedef enum _line_status_ {
/*-------------------------------------------------------------------------*/
/**
- @brief Convert a string to lowercase.
- @param s String to convert.
- @return ptr to statically allocated string.
-
- This function returns a pointer to a statically allocated string
- containing a lowercased version of the input string. Do not free
- or modify the returned string! Since the returned string is statically
- allocated, it will be modified at each function call (not re-entrant).
+ @brief Convert a string to lowercase.
+ @param in String to convert.
+ @param out Output buffer.
+ @param len Size of the out buffer.
+ @return ptr to the out buffer or NULL if an error occured.
+
+ This function convert a string into lowercase.
+ At most len - 1 elements of the input string will be converted.
*/
/*--------------------------------------------------------------------------*/
-
-static char strbuf[ASCIILINESZ+1];
-
-static char * strlwc(const char * s)
+static const char * strlwc(const char * in, char *out, unsigned len)
{
- int i ;
+ unsigned i ;
- if (s==NULL) return NULL ;
- memset(strbuf, 0, ASCIILINESZ+1);
+ if (in==NULL || out == NULL || len==0) return NULL ;
i=0 ;
- while (s[i] && i<ASCIILINESZ) {
- strbuf[i] = (char)tolower((int)s[i]);
+ while (in[i] != '\0' && i < len-1) {
+ out[i] = (char)tolower((int)in[i]);
i++ ;
}
- strbuf[ASCIILINESZ]=(char)0;
- return strbuf ;
+ out[i] = '\0';
+ return out ;
+}
+
+/*-------------------------------------------------------------------------*/
+/**
+ @brief Duplicate a string
+ @param s String to duplicate
+ @return Pointer to a newly allocated string, to be freed with free()
+
+ This is a replacement for strdup(). This implementation is provided
+ for systems that do not have it.
+ */
+/*--------------------------------------------------------------------------*/
+static char * xstrdup(const char * s)
+{
+ char * t ;
+ size_t len ;
+ if (!s)
+ return NULL ;
+
+ len = strlen(s) + 1 ;
+ t = (char*) malloc(len) ;
+ if (t) {
+ memcpy(t, s, len) ;
+ }
+ return t ;
+}
+
+/*-------------------------------------------------------------------------*/
+/**
+ @brief Remove blanks at the beginning and the end of a string.
+ @param str String to parse and alter.
+ @return unsigned New size of the string.
+ */
+/*--------------------------------------------------------------------------*/
+static unsigned strstrip(char * s)
+{
+ char *last = NULL ;
+ char *dest = s;
+
+ if (s==NULL) return 0;
+
+ last = s + strlen(s);
+ while (isspace((int)*s) && *s) s++;
+ while (last > s) {
+ if (!isspace((int)*(last-1)))
+ break ;
+ last -- ;
+ }
+ *last = (char)0;
+
+ memmove(dest,s,last - s + 1);
+ return last - s;
+}
+
+/*-------------------------------------------------------------------------*/
+/**
+ @brief Default error callback for iniparser: wraps `fprintf(stderr, ...)`.
+ */
+/*--------------------------------------------------------------------------*/
+static int default_error_callback(const char *format, ...)
+{
+ int ret;
+ va_list argptr;
+ va_start(argptr, format);
+ ret = vfprintf(stderr, format, argptr);
+ va_end(argptr);
+ return ret;
}
+static int (*iniparser_error_callback)(const char*, ...) = default_error_callback;
+
/*-------------------------------------------------------------------------*/
/**
- @brief Remove blanks at the beginning and the end of a string.
- @param s String to parse.
- @return ptr to statically allocated string.
-
- This function returns a pointer to a statically allocated string,
- which is identical to the input string, except that all blank
- characters at the end and the beg. of the string have been removed.
- Do not free or modify the returned string! Since the returned string
- is statically allocated, it will be modified at each function call
- (not re-entrant).
+ @brief Configure a function to receive the error messages.
+ @param errback Function to call.
+
+ By default, the error will be printed on stderr. If a null pointer is passed
+ as errback the error callback will be switched back to default.
*/
/*--------------------------------------------------------------------------*/
-static char * strstrip(const char * s)
+void iniparser_set_error_callback(int (*errback)(const char *, ...))
{
- char * last ;
-
- if (s==NULL) return NULL ;
-
- while (isspace((int)*s) && *s) s++;
- memset(strbuf, 0, ASCIILINESZ+1);
- strcpy(strbuf, s);
- last = strbuf + strlen(strbuf);
- while (last > strbuf) {
- if (!isspace((int)*(last-1)))
- break ;
- last -- ;
- }
- *last = (char)0;
- return (char*)strbuf ;
+ if (errback) {
+ iniparser_error_callback = errback;
+ } else {
+ iniparser_error_callback = default_error_callback;
+ }
}
/*-------------------------------------------------------------------------*/
@@ -118,7 +161,7 @@ static char * strstrip(const char * s)
This function returns -1 in case of error.
*/
/*--------------------------------------------------------------------------*/
-int iniparser_getnsec(dictionary * d)
+int iniparser_getnsec(const dictionary * d)
{
int i ;
int nsec ;
@@ -149,7 +192,7 @@ int iniparser_getnsec(dictionary * d)
This function returns NULL in case of error.
*/
/*--------------------------------------------------------------------------*/
-const char * iniparser_getsecname(dictionary * d, int n)
+const char * iniparser_getsecname(const dictionary * d, int n)
{
int i ;
int foundsec ;
@@ -184,7 +227,7 @@ const char * iniparser_getsecname(dictionary * d, int n)
purposes mostly.
*/
/*--------------------------------------------------------------------------*/
-void iniparser_dump(dictionary * d, FILE * f)
+void iniparser_dump(const dictionary * d, FILE * f)
{
int i ;
@@ -212,13 +255,11 @@ void iniparser_dump(dictionary * d, FILE * f)
It is Ok to specify @c stderr or @c stdout as output files.
*/
/*--------------------------------------------------------------------------*/
-void iniparser_dump_ini(dictionary * d, FILE * f)
+void iniparser_dump_ini(const dictionary * d, FILE * f)
{
- int i, j ;
- char keym[ASCIILINESZ+1];
- int nsec ;
- const char * secname ;
- int seclen ;
+ int i ;
+ int nsec ;
+ const char * secname ;
if (d==NULL || f==NULL) return ;
@@ -234,24 +275,126 @@ void iniparser_dump_ini(dictionary * d, FILE * f)
}
for (i=0 ; i<nsec ; i++) {
secname = iniparser_getsecname(d, i) ;
- seclen = (int)strlen(secname);
- fprintf(f, "\n[%s]\n", secname);
- sprintf(keym, "%s:", secname);
- for (j=0 ; j<d->size ; j++) {
- if (d->key[j]==NULL)
- continue ;
- if (!strncmp(d->key[j], keym, seclen+1)) {
- fprintf(f,
- "%-30s = %s\n",
- d->key[j]+seclen+1,
- d->val[j] ? d->val[j] : "");
- }
+ iniparser_dumpsection_ini(d, secname, f);
+ }
+ fprintf(f, "\n");
+ return ;
+}
+
+/*-------------------------------------------------------------------------*/
+/**
+ @brief Save a dictionary section to a loadable ini file
+ @param d Dictionary to dump
+ @param s Section name of dictionary to dump
+ @param f Opened file pointer to dump to
+ @return void
+
+ This function dumps a given section of a given dictionary into a loadable ini
+ file. It is Ok to specify @c stderr or @c stdout as output files.
+ */
+/*--------------------------------------------------------------------------*/
+void iniparser_dumpsection_ini(const dictionary * d, const char * s, FILE * f)
+{
+ int j ;
+ char keym[ASCIILINESZ+1];
+ int seclen ;
+
+ if (d==NULL || f==NULL) return ;
+ if (! iniparser_find_entry(d, s)) return ;
+
+ seclen = (int)strlen(s);
+ fprintf(f, "\n[%s]\n", s);
+ sprintf(keym, "%s:", s);
+ for (j=0 ; j<d->size ; j++) {
+ if (d->key[j]==NULL)
+ continue ;
+ if (!strncmp(d->key[j], keym, seclen+1)) {
+ fprintf(f,
+ "%-30s = %s\n",
+ d->key[j]+seclen+1,
+ d->val[j] ? d->val[j] : "");
}
}
fprintf(f, "\n");
return ;
}
+/*-------------------------------------------------------------------------*/
+/**
+ @brief Get the number of keys in a section of a dictionary.
+ @param d Dictionary to examine
+ @param s Section name of dictionary to examine
+ @return Number of keys in section
+ */
+/*--------------------------------------------------------------------------*/
+int iniparser_getsecnkeys(const dictionary * d, const char * s)
+{
+ int seclen, nkeys ;
+ char keym[ASCIILINESZ+1];
+ int j ;
+
+ nkeys = 0;
+
+ if (d==NULL) return nkeys;
+ if (! iniparser_find_entry(d, s)) return nkeys;
+
+ seclen = (int)strlen(s);
+ strlwc(s, keym, sizeof(keym));
+ keym[seclen] = ':';
+
+ for (j=0 ; j<d->size ; j++) {
+ if (d->key[j]==NULL)
+ continue ;
+ if (!strncmp(d->key[j], keym, seclen+1))
+ nkeys++;
+ }
+
+ return nkeys;
+
+}
+
+/*-------------------------------------------------------------------------*/
+/**
+ @brief Get the number of keys in a section of a dictionary.
+ @param d Dictionary to examine
+ @param s Section name of dictionary to examine
+ @param keys Already allocated array to store the keys in
+ @return The pointer passed as `keys` argument or NULL in case of error
+
+ This function queries a dictionary and finds all keys in a given section.
+ The keys argument should be an array of pointers which size has been
+ determined by calling `iniparser_getsecnkeys` function prior to this one.
+
+ Each pointer in the returned char pointer-to-pointer is pointing to
+ a string allocated in the dictionary; do not free or modify them.
+ */
+/*--------------------------------------------------------------------------*/
+const char ** iniparser_getseckeys(const dictionary * d, const char * s, const char ** keys)
+{
+ int i, j, seclen ;
+ char keym[ASCIILINESZ+1];
+
+ if (d==NULL || keys==NULL) return NULL;
+ if (! iniparser_find_entry(d, s)) return NULL;
+
+ seclen = (int)strlen(s);
+ strlwc(s, keym, sizeof(keym));
+ keym[seclen] = ':';
+
+ i = 0;
+
+ for (j=0 ; j<d->size ; j++) {
+ if (d->key[j]==NULL)
+ continue ;
+ if (!strncmp(d->key[j], keym, seclen+1)) {
+ keys[i] = d->key[j];
+ i++;
+ }
+ }
+
+ return keys;
+}
+
/*-------------------------------------------------------------------------*/
/**
@brief Get the string associated to a key
@@ -267,24 +410,27 @@ void iniparser_dump_ini(dictionary * d, FILE * f)
the dictionary, do not free or modify it.
*/
/*--------------------------------------------------------------------------*/
-const char * iniparser_getstring(dictionary * d, const char * key, const char * def)
+const char * iniparser_getstring(const dictionary * d, const char * key, const char * def)
{
- char * lc_key ;
+ const char * lc_key ;
+ const char * sval ;
+ char tmp_str[ASCIILINESZ+1];
if (d==NULL || key==NULL)
return def ;
- lc_key = strlwc(key);
- return dictionary_get(d, lc_key, def);
+ lc_key = strlwc(key, tmp_str, sizeof(tmp_str));
+ sval = dictionary_get(d, lc_key, def);
+ return sval ;
}
/*-------------------------------------------------------------------------*/
/**
- @brief Get the string associated to a key, convert to an int
+ @brief Get the string associated to a key, convert to an long int
@param d Dictionary to search
@param key Key string to look for
@param notfound Value to return in case of error
- @return integer
+ @return long integer
This function queries a dictionary for a key. A key as read from an
ini file is given as "section:key". If the key cannot be found,
@@ -305,13 +451,46 @@ const char * iniparser_getstring(dictionary * d, const char * key, const char *
Credits: Thanks to A. Becker for suggesting strtol()
*/
/*--------------------------------------------------------------------------*/
-int iniparser_getint(dictionary * d, const char * key, int notfound)
+long int iniparser_getlongint(const dictionary * d, const char * key, long int notfound)
{
- const char * str ;
+ const char * str ;
str = iniparser_getstring(d, key, INI_INVALID_KEY);
if (str==INI_INVALID_KEY) return notfound ;
- return (int)strtol(str, NULL, 0);
+ return strtol(str, NULL, 0);
+}
+
+
+/*-------------------------------------------------------------------------*/
+/**
+ @brief Get the string associated to a key, convert to an int
+ @param d Dictionary to search
+ @param key Key string to look for
+ @param notfound Value to return in case of error
+ @return integer
+
+ This function queries a dictionary for a key. A key as read from an
+ ini file is given as "section:key". If the key cannot be found,
+ the notfound value is returned.
+
+ Supported values for integers include the usual C notation
+ so decimal, octal (starting with 0) and hexadecimal (starting with 0x)
+ are supported. Examples:
+
+ "42" -> 42
+ "042" -> 34 (octal -> decimal)
+ "0x42" -> 66 (hexa -> decimal)
+
+ Warning: the conversion may overflow in various ways. Conversion is
+ totally outsourced to strtol(), see the associated man page for overflow
+ handling.
+
+ Credits: Thanks to A. Becker for suggesting strtol()
+ */
+/*--------------------------------------------------------------------------*/
+int iniparser_getint(const dictionary * d, const char * key, int notfound)
+{
+ return (int)iniparser_getlongint(d, key, notfound);
}
/*-------------------------------------------------------------------------*/
@@ -327,9 +506,9 @@ int iniparser_getint(dictionary * d, const char * key, int notfound)
the notfound value is returned.
*/
/*--------------------------------------------------------------------------*/
-double iniparser_getdouble(dictionary * d, const char * key, double notfound)
+double iniparser_getdouble(const dictionary * d, const char * key, double notfound)
{
- const char * str ;
+ const char * str ;
str = iniparser_getstring(d, key, INI_INVALID_KEY);
if (str==INI_INVALID_KEY) return notfound ;
@@ -368,10 +547,10 @@ double iniparser_getdouble(dictionary * d, const char * key, double notfound)
necessarily have to be 0 or 1.
*/
/*--------------------------------------------------------------------------*/
-int iniparser_getboolean(dictionary * d, const char * key, int notfound)
+int iniparser_getboolean(const dictionary * d, const char * key, int notfound)
{
- const char * c ;
- int ret ;
+ int ret ;
+ const char * c ;
c = iniparser_getstring(d, key, INI_INVALID_KEY);
if (c==INI_INVALID_KEY) return notfound ;
@@ -397,10 +576,7 @@ int iniparser_getboolean(dictionary * d, const char * key, int notfound)
of querying for the presence of sections in a dictionary.
*/
/*--------------------------------------------------------------------------*/
-int iniparser_find_entry(
- dictionary * ini,
- const char * entry
-)
+int iniparser_find_entry(const dictionary * ini, const char * entry)
{
int found=0 ;
if (iniparser_getstring(ini, entry, INI_INVALID_KEY)!=INI_INVALID_KEY) {
@@ -418,13 +594,14 @@ int iniparser_find_entry(
@return int 0 if Ok, -1 otherwise.
If the given entry can be found in the dictionary, it is modified to
- contain the provided value. If it cannot be found, -1 is returned.
+ contain the provided value. If it cannot be found, the entry is created.
It is Ok to set val to NULL.
*/
/*--------------------------------------------------------------------------*/
int iniparser_set(dictionary * ini, const char * entry, const char * val)
{
- return dictionary_set(ini, strlwc(entry), val) ;
+ char tmp_str[ASCIILINESZ+1];
+ return dictionary_set(ini, strlwc(entry, tmp_str, sizeof(tmp_str)), val) ;
}
/*-------------------------------------------------------------------------*/
@@ -439,12 +616,13 @@ int iniparser_set(dictionary * ini, const char * entry, const char * val)
/*--------------------------------------------------------------------------*/
void iniparser_unset(dictionary * ini, const char * entry)
{
- dictionary_unset(ini, strlwc(entry));
+ char tmp_str[ASCIILINESZ+1];
+ dictionary_unset(ini, strlwc(entry, tmp_str, sizeof(tmp_str)));
}
/*-------------------------------------------------------------------------*/
/**
- @brief Load a single line from an INI file
+ @brief Load a single line from an INI file
@param input_line Input line, may be concatenated multi-line input
@param section Output space to store section
@param key Output space to store key
@@ -457,34 +635,39 @@ static line_status iniparser_line(
char * section,
char * key,
char * value)
-{
+{
line_status sta ;
- char line[ASCIILINESZ+1];
- int len ;
+ char * line = NULL;
+ size_t len ;
- strcpy(line, strstrip(input_line));
- len = (int)strlen(line);
+ line = xstrdup(input_line);
+ len = strstrip(line);
sta = LINE_UNPROCESSED ;
if (len<1) {
/* Empty line */
sta = LINE_EMPTY ;
- } else if (line[0]=='#') {
+ } else if (line[0]=='#' || line[0]==';') {
/* Comment line */
- sta = LINE_COMMENT ;
+ sta = LINE_COMMENT ;
} else if (line[0]=='[' && line[len-1]==']') {
/* Section name */
sscanf(line, "[%[^]]", section);
- strcpy(section, strstrip(section));
- strcpy(section, strlwc(section));
+ strstrip(section);
+ strlwc(section, section, len);
sta = LINE_SECTION ;
} else if (sscanf (line, "%[^=] = \"%[^\"]\"", key, value) == 2
- || sscanf (line, "%[^=] = '%[^\']'", key, value) == 2
- || sscanf (line, "%[^=] = %[^;#]", key, value) == 2) {
- /* Usual key=value, with or without comments */
- strcpy(key, strstrip(key));
- strcpy(key, strlwc(key));
- strcpy(value, strstrip(value));
+ || sscanf (line, "%[^=] = '%[^\']'", key, value) == 2) {
+ /* Usual key=value with quotes, with or without comments */
+ strstrip(key);
+ strlwc(key, key, len);
+ /* Don't strip spaces from values surrounded with quotes */
+ sta = LINE_VALUE ;
+ } else if (sscanf (line, "%[^=] = %[^;#]", key, value) == 2) {
+ /* Usual key=value without quotes, with or without comments */
+ strstrip(key);
+ strlwc(key, key, len);
+ strstrip(value);
/*
* sscanf cannot handle '' or "" as empty values
* this is done here
@@ -501,14 +684,16 @@ static line_status iniparser_line(
* key=;
* key=#
*/
- strcpy(key, strstrip(key));
- strcpy(key, strlwc(key));
+ strstrip(key);
+ strlwc(key, key, len);
value[0]=0 ;
sta = LINE_VALUE ;
} else {
/* Generate syntax error */
sta = LINE_ERROR ;
}
+
+ free(line);
return sta ;
}
@@ -528,44 +713,33 @@ static line_status iniparser_line(
/*--------------------------------------------------------------------------*/
dictionary * iniparser_load(const char * ininame)
{
- char *buf;
FILE * in ;
- char *line;
- char *section;
- char *key;
- char *tmp;
- char *val;
+ char line [ASCIILINESZ+1] ;
+ char section [ASCIILINESZ+1] ;
+ char key [ASCIILINESZ+1] ;
+ char tmp [(ASCIILINESZ * 2) + 2] ;
+ char val [ASCIILINESZ+1] ;
int last=0 ;
int len ;
int lineno=0 ;
int errs=0;
- int ret;
+ int mem_err=0;
dictionary * dict ;
- if ((in=fopen(ininame, "r"))==NULL)
+ if ((in=fopen(ininame, "r"))==NULL) {
+ iniparser_error_callback("iniparser: cannot open %s\n", ininame);
return NULL ;
+ }
dict = dictionary_new(0) ;
if (!dict) {
fclose(in);
- errno = ENOMEM;
return NULL ;
}
- buf = malloc((ASCIILINESZ+1) * 5);
- if (buf == NULL) {
- errno = -ENOMEM;
- return NULL;
- }
- line = buf;
- section = line + ASCIILINESZ + 1;
- key = section + ASCIILINESZ + 1;
- tmp = key + ASCIILINESZ + 1;
- val = tmp + ASCIILINESZ + 1;
-
memset(line, 0, ASCIILINESZ);
memset(section, 0, ASCIILINESZ);
memset(key, 0, ASCIILINESZ);
@@ -575,18 +749,16 @@ dictionary * iniparser_load(const char * ininame)
while (fgets(line+last, ASCIILINESZ-last, in)!=NULL) {
lineno++ ;
len = (int)strlen(line)-1;
+ if (len<=0)
+ continue;
/* Safety check against buffer overflows */
- if (last > 0 && line[len]!='\n') {
-#if 0
- warning(anon_scope,
- "iniparser: input line too long in %s (%d)\n",
- ininame,
- lineno);
-#endif
+ if (line[len]!='\n' && !feof(in)) {
+ iniparser_error_callback(
+ "iniparser: input line too long in %s (%d)\n",
+ ininame,
+ lineno);
dictionary_del(dict);
fclose(in);
- free(buf);
- errno = EINVAL;
return NULL ;
}
/* Get rid of \n and spaces at end of line */
@@ -595,8 +767,11 @@ dictionary * iniparser_load(const char * ininame)
line[len]=0 ;
len-- ;
}
+ if (len < 0) { /* Line was entirely \n and/or spaces */
+ len = 0;
+ }
/* Detect multi-line */
- if (len >= 0 && line[len]=='\\') {
+ if (line[len]=='\\') {
/* Multi-line value */
last=len ;
continue ;
@@ -609,24 +784,20 @@ dictionary * iniparser_load(const char * ininame)
break ;
case LINE_SECTION:
- errs = dictionary_set(dict, section, NULL);
+ mem_err = dictionary_set(dict, section, NULL);
break ;
case LINE_VALUE:
sprintf(tmp, "%s:%s", section, key);
- errs = dictionary_set(dict, tmp, val) ;
+ mem_err = dictionary_set(dict, tmp, val);
break ;
- case LINE_ERROR:
-#if 0
- printf("iniparser: syntax error in %s (%d):\n",
- ininame,
- lineno);
- printf( "-> %s\n", line);
-
-#endif
-
- ret = EINVAL;
+ case LINE_ERROR:
+ iniparser_error_callback(
+ "iniparser: syntax error in %s (%d):\n-> %s\n",
+ ininame,
+ lineno,
+ line);
errs++ ;
break;
@@ -635,18 +806,16 @@ dictionary * iniparser_load(const char * ininame)
}
memset(line, 0, ASCIILINESZ);
last=0;
- if (errs<0) {
- ret = ENOMEM;
+ if (mem_err<0) {
+ iniparser_error_callback("iniparser: memory allocation failure\n");
break ;
}
}
- fclose(in);
- free(buf);
if (errs) {
dictionary_del(dict);
dict = NULL ;
- errno = ret;
}
+ fclose(in);
return dict ;
}
@@ -665,5 +834,3 @@ void iniparser_freedict(dictionary * d)
{
dictionary_del(d);
}
-
-/* vim: set ts=4 et sw=4 tw=75 */
diff --git a/lib/boilerplate/iniparser/iniparser.h b/lib/boilerplate/iniparser/iniparser.h
index d454cef34..37ff7b71b 100644
--- a/lib/boilerplate/iniparser/iniparser.h
+++ b/lib/boilerplate/iniparser/iniparser.h
@@ -3,22 +3,15 @@
/**
@file iniparser.h
@author N. Devillard
- @date Sep 2007
- @version 3.0
@brief Parser for ini files.
*/
/*--------------------------------------------------------------------------*/
-/*
- $Id: iniparser.h,v 1.24 2007-11-23 21:38:19 ndevilla Exp $
- $Revision: 1.24 $
-*/
-
#ifndef _INIPARSER_H_
#define _INIPARSER_H_
/*---------------------------------------------------------------------------
- Includes
+ Includes
---------------------------------------------------------------------------*/
#include <stdio.h>
@@ -34,12 +27,21 @@
#include "dictionary.h"
-/*---------------------------------------------------------------------------
- Macros
- ---------------------------------------------------------------------------*/
-/** For backwards compatibility only */
-#define iniparser_getstr(d, k) iniparser_getstring(d, k, NULL)
-#define iniparser_setstr iniparser_setstring
+#ifdef __cplusplus
+extern "C" {
+#endif
+
+/*-------------------------------------------------------------------------*/
+/**
+ @brief Configure a function to receive the error messages.
+ @param errback Function to call.
+
+ By default, the error will be printed on stderr. If a null pointer is passed
+ as errback the error callback will be switched back to default.
+ */
+/*--------------------------------------------------------------------------*/
+
+void iniparser_set_error_callback(int (*errback)(const char *, ...));
/*-------------------------------------------------------------------------*/
/**
@@ -60,7 +62,7 @@
*/
/*--------------------------------------------------------------------------*/
-int iniparser_getnsec(dictionary * d);
+int iniparser_getnsec(const dictionary * d);
/*-------------------------------------------------------------------------*/
@@ -78,7 +80,7 @@ int iniparser_getnsec(dictionary * d);
*/
/*--------------------------------------------------------------------------*/
-const char * iniparser_getsecname(dictionary * d, int n);
+const char * iniparser_getsecname(const dictionary * d, int n);
/*-------------------------------------------------------------------------*/
@@ -93,7 +95,22 @@ const char * iniparser_getsecname(dictionary * d, int n);
*/
/*--------------------------------------------------------------------------*/
-void iniparser_dump_ini(dictionary * d, FILE * f);
+void iniparser_dump_ini(const dictionary * d, FILE * f);
+
+/*-------------------------------------------------------------------------*/
+/**
+ @brief Save a dictionary section to a loadable ini file
+ @param d Dictionary to dump
+ @param s Section name of dictionary to dump
+ @param f Opened file pointer to dump to
+ @return void
+
+ This function dumps a given section of a given dictionary into a loadable ini
+ file. It is Ok to specify @c stderr or @c stdout as output files.
+ */
+/*--------------------------------------------------------------------------*/
+
+void iniparser_dumpsection_ini(const dictionary * d, const char * s, FILE * f);
/*-------------------------------------------------------------------------*/
/**
@@ -108,7 +125,36 @@ void iniparser_dump_ini(dictionary * d, FILE * f);
purposes mostly.
*/
/*--------------------------------------------------------------------------*/
-void iniparser_dump(dictionary * d, FILE * f);
+void iniparser_dump(const dictionary * d, FILE * f);
+
+/*-------------------------------------------------------------------------*/
+/**
+ @brief Get the number of keys in a section of a dictionary.
+ @param d Dictionary to examine
+ @param s Section name of dictionary to examine
+ @return Number of keys in section
+ */
+/*--------------------------------------------------------------------------*/
+int iniparser_getsecnkeys(const dictionary * d, const char * s);
+
+/*-------------------------------------------------------------------------*/
+/**
+ @brief Get the number of keys in a section of a dictionary.
+ @param d Dictionary to examine
+ @param s Section name of dictionary to examine
+ @param keys Already allocated array to store the keys in
+ @return The pointer passed as `keys` argument or NULL in case of error
+
+ This function queries a dictionary and finds all keys in a given section.
+ The keys argument should be an array of pointers which size has been
+ determined by calling `iniparser_getsecnkeys` function prior to this one.
+
+ Each pointer in the returned char pointer-to-pointer is pointing to
+ a string allocated in the dictionary; do not free or modify them.
+ */
+/*--------------------------------------------------------------------------*/
+const char ** iniparser_getseckeys(const dictionary * d, const char * s, const char ** keys);
+
/*-------------------------------------------------------------------------*/
/**
@@ -125,7 +171,7 @@ void iniparser_dump(dictionary * d, FILE * f);
the dictionary, do not free or modify it.
*/
/*--------------------------------------------------------------------------*/
-const char * iniparser_getstring(dictionary * d, const char * key, const char * def);
+const char * iniparser_getstring(const dictionary * d, const char * key, const char * def);
/*-------------------------------------------------------------------------*/
/**
@@ -154,7 +200,35 @@ const char * iniparser_getstring(dictionary * d, const char * key, const char *
Credits: Thanks to A. Becker for suggesting strtol()
*/
/*--------------------------------------------------------------------------*/
-int iniparser_getint(dictionary * d, const char * key, int notfound);
+int iniparser_getint(const dictionary * d, const char * key, int notfound);
+
+/*-------------------------------------------------------------------------*/
+/**
+ @brief Get the string associated to a key, convert to an long int
+ @param d Dictionary to search
+ @param key Key string to look for
+ @param notfound Value to return in case of error
+ @return integer
+
+ This function queries a dictionary for a key. A key as read from an
+ ini file is given as "section:key". If the key cannot be found,
+ the notfound value is returned.
+
+ Supported values for integers include the usual C notation
+ so decimal, octal (starting with 0) and hexadecimal (starting with 0x)
+ are supported. Examples:
+
+ - "42" -> 42
+ - "042" -> 34 (octal -> decimal)
+ - "0x42" -> 66 (hexa -> decimal)
+
+ Warning: the conversion may overflow in various ways. Conversion is
+ totally outsourced to strtol(), see the associated man page for overflow
+ handling.
+ */
+/*--------------------------------------------------------------------------*/
+long int iniparser_getlongint(const dictionary * d, const char * key, long int notfound);
+
/*-------------------------------------------------------------------------*/
/**
@@ -169,7 +243,7 @@ int iniparser_getint(dictionary * d, const char * key, int notfound);
the notfound value is returned.
*/
/*--------------------------------------------------------------------------*/
-double iniparser_getdouble(dictionary * d, const char * key, double notfound);
+double iniparser_getdouble(const dictionary * d, const char * key, double notfound);
/*-------------------------------------------------------------------------*/
/**
@@ -203,7 +277,7 @@ double iniparser_getdouble(dictionary * d, const char * key, double notfound);
necessarily have to be 0 or 1.
*/
/*--------------------------------------------------------------------------*/
-int iniparser_getboolean(dictionary * d, const char * key, int notfound);
+int iniparser_getboolean(const dictionary * d, const char * key, int notfound);
/*-------------------------------------------------------------------------*/
@@ -212,17 +286,16 @@ int iniparser_getboolean(dictionary * d, const char * key, int notfound);
@param ini Dictionary to modify.
@param entry Entry to modify (entry name)
@param val New value to associate to the entry.
- @return int 0 if Ok, -1 otherwise.
+ @return int 0 if Ok, -1 otherwise.
If the given entry can be found in the dictionary, it is modified to
- contain the provided value. If it cannot be found, -1 is returned.
+ contain the provided value. If it cannot be found, the entry is created.
It is Ok to set val to NULL.
*/
/*--------------------------------------------------------------------------*/
-int iniparser_setstring(dictionary * ini, const char * entry, const char * val);
-
int iniparser_set(dictionary * ini, const char * entry, const char * val);
+
/*-------------------------------------------------------------------------*/
/**
@brief Delete an entry in a dictionary
@@ -247,7 +320,7 @@ void iniparser_unset(dictionary * ini, const char * entry);
of querying for the presence of sections in a dictionary.
*/
/*--------------------------------------------------------------------------*/
-int iniparser_find_entry(dictionary * ini, const char * entry) ;
+int iniparser_find_entry(const dictionary * ini, const char * entry) ;
/*-------------------------------------------------------------------------*/
/**
@@ -278,4 +351,8 @@ dictionary * iniparser_load(const char * ininame);
/*--------------------------------------------------------------------------*/
void iniparser_freedict(dictionary * d);
+#ifdef __cplusplus
+}
+#endif
+
#endif
--
2.26.2
^ permalink raw reply related [flat|nested] 5+ messages in thread
* Re: [PATCH v2] lib/boilerplate/iniparser: Allow building with GCC 10.2 2020101
2020-11-12 11:45 ` [PATCH v2] lib/boilerplate/iniparser: Allow building with GCC 10.2 2020101 florian.bezdeka
@ 2020-11-12 16:19 ` Jan Kiszka
0 siblings, 0 replies; 5+ messages in thread
From: Jan Kiszka @ 2020-11-12 16:19 UTC (permalink / raw)
To: Bezdeka, Florian (T RDA IOT SES-DE), xenomai
On 12.11.20 12:45, Bezdeka, Florian (T RDA IOT SES-DE) wrote:
> From: Florian Bezdeka <florian.bezdeka@siemens.com>
>
> Updating to upstream revision f858275f7f307eecba84c2f5429483f9f28007f8.
>
> The reason for updating was the following compiler error when trying
> to compile with GCC 10.2 10.2.1 20201016. As it turned out the problem
> was already addressed upstream:
>
> iniparser/iniparser.c: In function ‘iniparser_load’:
> iniparser/iniparser.c:616:13: error: ‘sprintf’ arguments 3, 4 may
> overlap destination object ‘buf’ [-Werror=restrict]
> 616 | sprintf(tmp, "%s:%s", section, key);
> | ^~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
>
> I reviewed especially the API changes. Most of them are cleanups only
> but two things should be pointed out:
>
> - The type of the size field of struct _dictionary_ changed from int
> to ssize_t. The only user of this struct is
> lib/analogy/calibration.c which uses this structure for internal
> things only. It is never exposed to any public API so updating is
> OK and fully backward compatible.
>
> - dictionary_new changed its signature
> from dictionary_new(int size)
> to dictionary_new(size_t size).
> This function is not part of any public API. So updating does not
> break backward compatibility.
>
> Signed-off-by: Florian Bezdeka <florian.bezdeka@siemens.com>
> ---
> lib/boilerplate/iniparser/dictionary.c | 409 ++++++++++----------
> lib/boilerplate/iniparser/dictionary.h | 43 ++-
> lib/boilerplate/iniparser/iniparser.c | 491 +++++++++++++++++--------
> lib/boilerplate/iniparser/iniparser.h | 131 +++++--
> 4 files changed, 646 insertions(+), 428 deletions(-)
>
Thanks, applied to next (after sorting out our Siemens-local mail
transfer problem...).
Jan
--
Siemens AG, T RDA IOT
Corporate Competence Center Embedded Linux
^ permalink raw reply [flat|nested] 5+ messages in thread
end of thread, other threads:[~2020-11-12 16:19 UTC | newest]
Thread overview: 5+ messages (download: mbox.gz / follow: Atom feed)
-- links below jump to the message on this page --
2020-11-11 14:59 [PATCH] lib/boilerplate: Allow building with GCC 10.2 florian.bezdeka
2020-11-12 6:58 ` Jan Kiszka
2020-11-12 8:49 ` florian.bezdeka
2020-11-12 11:45 ` [PATCH v2] lib/boilerplate/iniparser: Allow building with GCC 10.2 2020101 florian.bezdeka
2020-11-12 16:19 ` Jan Kiszka
This is an external index of several public inboxes,
see mirroring instructions on how to clone and mirror
all data and code used by this external index.