Skip to content

folly/cli: eliminate double scan in cli_parse_args_from_content EOF path #2647

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

Open
darion-yaphet wants to merge 1 commit into
base: main
Choose a base branch
from
+22 −20
Open

folly/cli: eliminate double scan in cli_parse_args_from_content EOF path #2647

Changes from all commits
Commits
Show all changes
1 commit
Select commit
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
42 changes: 22 additions & 20 deletions folly/cli/Args.cpp
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -144,6 +144,11 @@ cli_parse_result cli_parse_args_from_content(std::string_view content) {
size_t token_start_col = 0;
size_t token_end_line = 0;
size_t token_end_col = 0;
// Byte offset in content of the last non-whitespace character seen in the
// current token. Updated in every branch that also updates token_end_line/
// col so that the EOF path can compute length in O(1) without a backward
// scan of content.
size_t token_end_offset = 0;

// Helper to finalize and push a token
auto push_token = [&](size_t end_pos) {
Expand Down Expand Up @@ -187,6 +192,7 @@ cli_parse_result cli_parse_args_from_content(std::string_view content) {
in_single_quotes = false;
token_end_line = line;
token_end_col = col;
token_end_offset = i;
col++;
continue;
}
Expand All @@ -195,6 +201,7 @@ cli_parse_result cli_parse_args_from_content(std::string_view content) {
chars_in_token++;
token_end_line = line;
token_end_col = col;
token_end_offset = i;
if (c == '\n') {
line++;
col = 1;
Expand All @@ -214,6 +221,7 @@ cli_parse_result cli_parse_args_from_content(std::string_view content) {
}
token_end_line = line;
token_end_col = col;
token_end_offset = i;
switch (c) {
case 'n':
current += '\n';
Expand Down Expand Up @@ -265,6 +273,7 @@ cli_parse_result cli_parse_args_from_content(std::string_view content) {
}
token_end_line = line;
token_end_col = col;
token_end_offset = i;
escaped = true;
col++;
continue;
Expand All @@ -286,6 +295,7 @@ cli_parse_result cli_parse_args_from_content(std::string_view content) {
}
token_end_line = line;
token_end_col = col;
token_end_offset = i;
in_single_quotes = true;
quote_start_line = line;
quote_start_col = col;
Expand All @@ -303,6 +313,7 @@ cli_parse_result cli_parse_args_from_content(std::string_view content) {
}
token_end_line = line;
token_end_col = col;
token_end_offset = i;
if (in_double_quotes) {
in_double_quotes = false;
} else {
Expand Down Expand Up @@ -337,6 +348,7 @@ cli_parse_result cli_parse_args_from_content(std::string_view content) {
if (!cli_is_space(c)) {
token_end_line = line;
token_end_col = col;
token_end_offset = i;
}

// Track @ prefix: only meaningful when outside quotes
Expand Down Expand Up @@ -381,29 +393,19 @@ cli_parse_result cli_parse_args_from_content(std::string_view content) {
}

if (!current.empty()) {
// Strip trailing whitespace from the last token
// This handles cases where the file ends with whitespace inside quotes
// or simply trailing whitespace at the end
while (!current.empty()) {
char c = current.back();
if (cli_is_space(c)) {
current.pop_back();
} else {
break;
}
// Strip trailing whitespace from the token value. This handles the case
// where a token accumulates spaces while inside double-quotes that were
// then left open until EOF (which would have been caught above), or any
// other edge case that leaves spaces at the tail of `current`.
while (!current.empty() && cli_is_space(current.back())) {
current.pop_back();
}

if (!current.empty()) {
size_t length = content.size() - token_start;
// Find the actual end of the token (excluding trailing whitespace)
for (size_t i = content.size(); i > token_start; --i) {
char c = content[i - 1];
if (!cli_is_space(c)) {
length = i - token_start;
break;
}
}

// token_end_offset is the byte index of the last non-whitespace character
// in this token, maintained throughout the main scan loop. Using it
// avoids a second backward scan over `content` to recompute length.
size_t length = token_end_offset + 1 - token_start;
parse_result.args.push_back(
{std::move(current),
token_start,
Expand Down