Skip to content

fix trailing space stripping for different encodings #10

New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Merged
sachintu47 merged 3 commits into from
May 11, 2026
Merged

fix trailing space stripping for different encodings #10

Changes from all commits
Commits
Show all changes
3 commits
Select commit Hold shift + click to select a range
00c672f
fix trailing space stripping for different encodings
sachintu47 May 11, 2026
ec245f2
disable stripping for multi-byte encodings and restrict binary scan t…
sachintu47 May 11, 2026
588f405
augment review fixes
sachintu47 May 11, 2026
File filter

Filter by extension

Filter by extension
Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
53 changes: 48 additions & 5 deletions src/libdio.c
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -17,6 +17,7 @@
#include <_Nascii.h>
#include <unistd.h>
#include <cjson/cJSON.h>
#include <_Ccsid.h>

#define _OPEN_SYS_EXT
#include <sys/ps.h>
Expand All @@ -28,6 +29,22 @@

const struct s99_rbx s99rbxtemplate = {"S99RBX",S99RBXVR,{0,1,0,0,0,0,0},0,0,0};

static int get_space_char(int ccsid) {
if (ccsid == 0) {
return 0x40; // Default to EBCDIC for untagged datasets
}
if (ccsid > 0) {
__csType cs = __CcsidType(ccsid);
if (cs == _CSTYPE_ASCII || cs == _CSTYPE_UTF8) {
return 0x20; // ASCII space
}
if (cs == _CSTYPE_EBCDIC) {
return 0x40; // EBCDIC space
}
}
return -1; // Default to disable trimming for unknown/multi-byte encodings
}

void dbgmsg(struct DFILE* dfile, const char* format, ...)
{
if (!dfile->debug) {
Expand Down Expand Up @@ -703,9 +720,9 @@ static enum DIOERR read_dataset_internal(struct DFILE* dfile)
memcpy(&dfile->buffer[difile->cur_read_offset], &reclen, sizeof(reclen));
difile->cur_read_offset += sizeof(reclen);
}
memcpy(&dfile->buffer[difile->cur_read_offset], record, bytes_to_copy);
if (!is_binary)
isbinary = is_binary(&dfile->buffer[difile->cur_read_offset], bytes_to_copy);
memcpy(&dfile->buffer[difile->cur_read_offset], record, rc);
if (!isbinary)
isbinary = is_binary(&dfile->buffer[difile->cur_read_offset], rc);
#ifdef DEBUG
printf("%5.5u <%*.*s>\n", reclen, reclen, reclen, record);
#endif
Expand Down Expand Up @@ -1144,15 +1161,32 @@ int write_dataset_to_temp_file(struct DFILE *dfile, char *tempname,
struct f_cnvrt req = {SETCVTOFF, 0, 0};
fcntl(temp_fd, F_CONTROL_CVT, &req);

int space_char = get_space_char(dfile->ccsid);
Comment thread
sachintu47 marked this conversation as resolved.
int length_prefix = has_length_prefix(dfile->recfm);
int i = 0;
char *data = dfile->buffer;
if (length_prefix) {
uint16_t reclen;
while (i < dfile->bufflen) {
if (i + sizeof(reclen) > dfile->bufflen) {
close_dataset(dfile);
close(temp_fd);
return 1; // Corrupt buffer: header truncated
}
memcpy(&reclen, &data[i], sizeof(reclen));
i += sizeof(reclen);
if (write(temp_fd, &data[i], reclen) != reclen) {
int actual_len = reclen;
if (i + actual_len > dfile->bufflen) {
close_dataset(dfile);
close(temp_fd);
return 1; // Corrupt buffer: record truncated
}
if (!force_binary && space_char != -1) {
while (actual_len > 0 && data[i + actual_len - 1] == space_char) {
Comment thread
sachintu47 marked this conversation as resolved.
Comment thread
sachintu47 marked this conversation as resolved.
actual_len--;
}
}
if (write(temp_fd, &data[i], actual_len) != actual_len) {
close_dataset(dfile);
close(temp_fd);
return 1;
Expand All @@ -1167,7 +1201,16 @@ int write_dataset_to_temp_file(struct DFILE *dfile, char *tempname,
}
} else {
while (i < dfile->bufflen) {
if (write(temp_fd, &data[i], dfile->reclen) != dfile->reclen) {
int actual_len = dfile->reclen;
if (i + actual_len > dfile->bufflen) {
Copy link
Copy Markdown

@augmentcode augmentcode Bot May 11, 2026 •
edited
Loading

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

In the fixed-length (no length prefix) path, clamping actual_len to the remaining buffer will silently write a partial final record if bufflen isn’t a multiple of reclen, which can mask buffer corruption and produce output that downstream code can’t parse reliably. Should this case be treated as an error (similar to the length-prefixed branch) instead of emitting a truncated record?

Severity: medium

Fix This in Augment

🤖 Was this useful? React with 👍 or 👎, or 🚀 if it prevented an incident/outage.

actual_len = dfile->bufflen - i; // Clamp to remaining buffer
}
if (!force_binary && space_char != -1) {
while (actual_len > 0 && data[i + actual_len - 1] == space_char) {
Comment thread
sachintu47 marked this conversation as resolved.
actual_len--;
}
}
if (write(temp_fd, &data[i], actual_len) != actual_len) {
close_dataset(dfile);
close(temp_fd);
return 1;
Expand Down
Loading