fix validation of long strings

We were failing silently if string length validation failed. With this PR,
issue a warning and let long strings though.

Also, add a `-w` flag (show warnings) to `dcm-dump` and `dcm-getframe`.

Thanks @arngaillard

See #105

Author: jcupitt

CHANGELOG.md

@@ -2,6 +2,8 @@
 
 * fix one-byte overread into struct padding [bgilbert]
 * support single-frame DICOM images and allow BitsStored > 8 [tokyovigilante]
+* fix error handling for string values over max length [arngaillard]
+* add `-w` (show warnings) to dcm-dump and dcm-getframe
 
 ## 1.2.0, 09/04/2025
 

doc/source/usage.rst

@@ -55,6 +55,11 @@ using a standard C type (e.g,. VR ``"US"`` has type ``uint16_t`` and VR
 ``"UI"`` has type ``char *``) and additional value constraints may be
 checked at runtime (e.g., the maximal capacity of a character string).
 
+When reading DICOM files, a warning will be issued for character strings which
+are too long, but they will still be passed unmodified to the caller. Readers
+must not crash if strings are longer than the maximum value in the DICOM
+specification.
+
 The VR must be appropriate for the tag. Use :c:func:`dcm_vr_from_tag()` to
 find the set of allowed VRs for a tag. Use :c:func:`dcm_is_valid_vr_for_tag()`
 to check if a VR is allowed for a tag.

src/dicom-data.c

@@ -340,14 +340,12 @@ bool dcm_element_get_value_string(DcmError **error,
 static bool element_check_capacity(DcmError **error,
                                    DcmElement *element, uint32_t capacity)
 {
-    uint32_t i;
-
     bool was_assigned = element->assigned;
 
     // we have to turn on "assigned" for this func so we can read out values
     element->assigned = true;
 
-    for (i = 0; i < element->vm; i++) {
+    for (uint32_t i = 0; i < element->vm; i++) {
         const char *value;
         if (!dcm_element_get_value_string(error, element, i, &value)) {
             element->assigned = was_assigned;
@@ -402,9 +400,9 @@ static bool dcm_element_validate(DcmError **error, DcmElement *element)
     if (vr_class == DCM_VR_CLASS_STRING_MULTI ||
         vr_class == DCM_VR_CLASS_STRING_SINGLE) {
         uint32_t capacity = dcm_dict_vr_capacity(element->vr);
-        if (!element_check_capacity(error, element, capacity)) {
-            return false;
-        }
+
+        // this only checks and issues a warning, it does not set error
+        (void) element_check_capacity(error, element, capacity);
     }
 
     element->assigned = true;

tools/dcm-dump.1.in

@@ -4,13 +4,14 @@
 dcm-dump \- print metadata content of DICOM PS3.10 file to standard output
 
 .SH SYNOPSIS
-.BR "dcm-dump " [ -v "]
+.BR "dcm-dump " [ -hViw "]
 .IR file
 
 .SH DESCRIPTION
 Print metadata content of DICOM PS3.10 file to standard output.
 
 .SH OPTIONS
+
 .TP
 .B -h
 Display help message (usage summary) and exit.
@@ -20,8 +21,12 @@ Display help message (usage summary) and exit.
 Display version and exit.
 
 .TP
-.B -v
-Increase logging verbosity to INFO.
+.B -i
+Show info messages.
+
+.TP
+.B -w
+Show warning messages.
 
 .SH EXIT STATUS
 .B dcm-dump

tools/dcm-dump.c

@@ -7,35 +7,39 @@
 #include <dicom/dicom.h>
 
 
-static const char usage[] = "usage: dcm-dump [-v] [-V] [-h] FILE_PATH ...";
+static const char usage[] = "usage: dcm-dump [-hViw] FILE_PATH ...";
 
 
 int main(int argc, char *argv[])
 {
-    int i, c;
-
-    while ((c = dcm_getopt(argc, argv, "h?Vv")) != -1) {
+    int c;
+    while ((c = dcm_getopt(argc, argv, "h?Vviw")) != -1) {
         switch (c) {
             case 'h':
             case '?':
                 printf("%s\n", usage);
                 return EXIT_SUCCESS;
 
-            case 'v':
+            case 'V':
                 printf("%s\n", dcm_get_version());
                 return EXIT_SUCCESS;
 
-            case 'V':
+            case 'v':
+            case 'i':
                 dcm_log_set_level(DCM_LOG_INFO);
                 break;
 
+            case 'w':
+                dcm_log_set_level(DCM_LOG_WARNING);
+                break;
+
             case '#':
             default:
                 return EXIT_FAILURE;
         }
     }
 
-    for (i = dcm_optind; i < argc; i++) {
+    for (int i = dcm_optind; i < argc; i++) {
         DcmError *error = NULL;
         DcmFilehandle *filehandle = NULL;
 
@@ -45,8 +49,8 @@ int main(int argc, char *argv[])
             dcm_error_print(error);
             dcm_error_clear(&error);
             return EXIT_FAILURE;
-        }  
-        
+        }
+
         if (!dcm_filehandle_print(&error, filehandle)) {
             dcm_error_print(error);
             dcm_error_clear(&error);

tools/dcm-getframe.1.in

@@ -4,18 +4,14 @@
 dcm-getframe \- print a frame from a DICOM PS3.10 file to standard output
 
 .SH SYNOPSIS
-.BR "dcm-getframe " [ -v "]
+.BR "dcm-getframe " [ -hViw "]
 .IR file
 .IR frame-number
 
 .SH DESCRIPTION
 Print a frame from a DICOM PS3.10 file to standard output.
 
 .SH OPTIONS
-.TP
-.B -o OUTPUT-FILENAME
-Write frame to OUTPUT-FILENAME. By default, the frame will be written to
-stdout.
 
 .TP
 .B -h
@@ -26,8 +22,17 @@ Display help message (usage summary) and exit.
 Display version and exit.
 
 .TP
-.B -v
-Increase logging verbosity to INFO.
+.B -o OUTPUT-FILENAME
+Write frame to OUTPUT-FILENAME. By default, the frame will be written to
+stdout.
+
+.TP
+.B -i
+Show info messages.
+
+.TP
+.B -w
+Show warning messages.
 
 .SH EXIT STATUS
 .B dcm-getframe

tools/dcm-getframe.c

@@ -11,16 +11,15 @@
 
 
 static const char usage[] = "usage: "
-    "dcm-getframe [-v] [-V] [-h] [-o OUTPUT-FILE] FILE_PATH FRAME_NUMBER";
+    "dcm-getframe [-hviw] [-o OUTPUT-FILE] FILE_PATH FRAME_NUMBER";
 
 
 int main(int argc, char *argv[])
 {
     char *output_file = NULL;
 
     int c;
-
-    while ((c = dcm_getopt(argc, argv, "h?Vvo:")) != -1) {
+    while ((c = dcm_getopt(argc, argv, "h?vViwo:")) != -1) {
         switch (c) {
             case 'h':
             case '?':
@@ -32,9 +31,14 @@ int main(int argc, char *argv[])
                 return EXIT_SUCCESS;
 
             case 'V':
+            case 'i':
                 dcm_log_set_level(DCM_LOG_INFO);
                 break;
 
+            case 'w':
+                dcm_log_set_level(DCM_LOG_WARNING);
+                break;
+
             case 'o':
                 output_file = dcm_optarg;
                 break;