Files
ik_llama.cpp/common/jinja/string.cpp
firecoperana ab1d74074b common : introduce composable PEG parser combinators for chat parsing and new jinja template engine (#1369)
---------

Co-authored-by: Piotr Wilkin <piotr.wilkin@syndatis.com>

common : add nemotron 3 parsing (#18077)

common : add parser for ministral/mistral large 3/devstral 2 (#17713)

common : default content to an empty string (#18485)

chat: make tool description and parameters optional per OpenAI spec (#18478)

Per the OpenAI API specification, both 'description' and 'parameters'
fields in tool function definitions are optional. Previously, the parser
would throw an exception if these fields were missing.

Attempts to fix #17667

common : implement new jinja template engine (#18462)
---------

Co-authored-by: Alde Rojas <hello@alde.dev>
Co-authored-by: Sigbjørn Skjæret <sigbjorn.skjaeret@scala.com>
Co-authored-by: Georgi Gerganov <ggerganov@gmail.com>

jinja: correct member access rule (#18905)

jinja : fix lexing of float literals with sign (#18901)

jinja : add missing tojson filter for bool (#18900)

jinja : attribute support for join, map and sort (#18883)

jinja : fix object item order (and properly implement dictsort) (#18904)

tests : add test-jinja -py option for cross-checking (#18906)

Co-authored-by: Sigbjørn Skjæret <sigbjorn.skjaeret@scala.com>

---------

Co-authored-by: Sigbjørn Skjæret <sigbjorn.skjaeret@scala.com>

ci : run test-jinja -py on high perf [no ci] (#18916)

jinja : fix undefined keys and attributes and int/float as bool (#18924)

jinja: support none|string (#18995)

Co-authored-by: Sigbjørn Skjæret <sigbjorn.skjaeret@scala.com>

Co-authored-by: Sigbjørn Skjæret <sigbjorn.skjaeret@scala.com>

---------

Co-authored-by: Sigbjørn Skjæret <sigbjorn.skjaeret@scala.com>

jinja : implement mixed type object keys (#18955)

---------

Co-authored-by: Xuan Son Nguyen <son@huggingface.co>

jinja : undefined should be treated as sequence/iterable (return string/array) by filters/tests (#19147)

`tojson` is not a supported `undefined` filter

keep it DRY and fix some types

jinja : do not pass empty tools and add some none filters (#19176)

jinja : add unordered_map include to value.h [no ci] (#19205)

jinja : add missing 'in' test to template engine (#19004) (#19239)

The jinja template parser was missing the 'in' test from
global_builtins(), causing templates using reject("in", ...),
select("in", ...), or 'x is in(y)' to fail with
"selectattr: unknown test 'in'".

This broke tool-calling for Qwen3-Coder and any other model
whose chat template uses the 'in' test.

Added test_is_in supporting array, string, and object containment
checks, mirroring the existing 'in' operator logic in runtime.cpp.

Includes test cases for all three containment types plus
reject/select filter usage.

Co-Authored-By: Claude Opus 4.5 <noreply@anthropic.com>

---------

Co-authored-by: Sid Mohan <sidmohan0@users.noreply.github.com>
Co-authored-by: Claude Opus 4.5 <noreply@anthropic.com>
Co-authored-by: Xuan Son Nguyen <son@huggingface.co>

Add Jinja support for "indent" string filter (#19529)

Co-authored-by: Sigbjørn Skjæret <sigbjorn.skjaeret@scala.com>

Co-authored-by: Sigbjørn Skjæret <sigbjorn.skjaeret@scala.com>

---------

Co-authored-by: Sigbjørn Skjæret <sigbjorn.skjaeret@scala.com>

add vendor

refactor chat

server : support preserving reasoning_content in assistant message (#18994)

chat : fix translategemma crash on common_chat_format_example (#19019)

chat: fix language input for translategemma (#19052)

Co-authored-by: Aldehir Rojas <hello@alde.dev>

---------

Co-authored-by: Aldehir Rojas <hello@alde.dev>

chat: fix case where template accepts type content only (#19419)

mtmd : chat : Fix extra \n between text and media marker (#19595)

Thanks to @tugot17 for detecting and reporting the issue.

For vision models (e.g. LFM2.5-VL-1.6B and Qwen/Qwen3-VL-4B-Instruct) `llama-mtmd-cli` produces identical output to HF implementation.

However `llama-server` doesn't. I traced it down to extra newline
inserted after `<__media__>`.

This happens in `to_json_oaicompat`, that treats media markers as text
and joins all parts with `\n` separator.

PR introduces new type `media_marker` and uses it for media markers.
Extra logic is added to prevent insertion of newlines before and after
media markers.

With this change number of input tokens is identical to HF
implementation and as a result the output is also identical.

I explored other ways to address the issue
* remove completely `\n` between text parts in `to_json_oaicompat`
* merge text messages in server-common.cpp before sending them to `to_json_oaicompat`

Please propose alternative ways of fixing this issue.

Co-authored-by: Piotr Wilkin (ilintar) <piotr.wilkin@syndatis.com>

---------

Co-authored-by: Piotr Wilkin (ilintar) <piotr.wilkin@syndatis.com>

common : merge qwen3-coder and nemotron nano 3 parsers (#19765)

common : fix improper trimming in XML parser on complete message (#19805)

Co-authored-by: Jules LEIDELINGER <11395311+julio75012@users.noreply.github.com>

jinja: correct stats for tojson and string filters (#19785)

jinja : correct default size for string slices (#19913)

common : handle unicode during partial json parsing (#16526)

common : fix json schema with '\' in literals (#17307)

add back qwen_coder_xml and mirothinker

Co-authored-by: Aldehir Rojas <hello@alde.dev>
2026-03-09 11:03:33 +01:00

214 lines
5.3 KiB
C++

#include "jinja/string.h"
#include "jinja/value.h"
#include <algorithm>
#include <functional>
#include <optional>
#include <sstream>
#include <string>
#include <vector>
namespace jinja {
//
// string_part
//
bool string_part::is_uppercase() const {
for (char c : val) {
if (std::islower(static_cast<unsigned char>(c))) {
return false;
}
}
return true;
}
bool string_part::is_lowercase() const {
for (char c : val) {
if (std::isupper(static_cast<unsigned char>(c))) {
return false;
}
}
return true;
}
//
// string
//
void string::mark_input() {
for (auto & part : parts) {
part.is_input = true;
}
}
std::string string::str() const {
if (parts.size() == 1) {
return parts[0].val;
}
std::ostringstream oss;
for (const auto & part : parts) {
oss << part.val;
}
return oss.str();
}
size_t string::length() const {
size_t len = 0;
for (const auto & part : parts) {
len += part.val.length();
}
return len;
}
void string::hash_update(hasher & hash) const noexcept {
for (const auto & part : parts) {
hash.update(part.val.data(), part.val.length());
}
}
bool string::all_parts_are_input() const {
for (const auto & part : parts) {
if (!part.is_input) {
return false;
}
}
return true;
}
bool string::is_uppercase() const {
for (const auto & part : parts) {
if (!part.is_uppercase()) {
return false;
}
}
return true;
}
bool string::is_lowercase() const {
for (const auto & part : parts) {
if (!part.is_lowercase()) {
return false;
}
}
return true;
}
// mark this string as input if other has ALL parts as input
void string::mark_input_based_on(const string & other) {
if (other.all_parts_are_input()) {
for (auto & part : parts) {
part.is_input = true;
}
}
}
string string::append(const string & other) {
for (const auto & part : other.parts) {
parts.push_back(part);
}
return *this;
}
// in-place transformation
using transform_fn = std::function<std::string(const std::string&)>;
static string apply_transform(string & self, const transform_fn & fn) {
for (auto & part : self.parts) {
part.val = fn(part.val);
}
return self;
}
string string::uppercase() {
return apply_transform(*this, [](const std::string & s) {
std::string res = s;
std::transform(res.begin(), res.end(), res.begin(), ::toupper);
return res;
});
}
string string::lowercase() {
return apply_transform(*this, [](const std::string & s) {
std::string res = s;
std::transform(res.begin(), res.end(), res.begin(), ::tolower);
return res;
});
}
string string::capitalize() {
return apply_transform(*this, [](const std::string & s) {
if (s.empty()) return s;
std::string res = s;
res[0] = ::toupper(static_cast<unsigned char>(res[0]));
std::transform(res.begin() + 1, res.end(), res.begin() + 1, ::tolower);
return res;
});
}
string string::titlecase() {
return apply_transform(*this, [](const std::string & s) {
std::string res = s;
bool capitalize_next = true;
for (char &c : res) {
if (isspace(static_cast<unsigned char>(c))) {
capitalize_next = true;
} else if (capitalize_next) {
c = ::toupper(static_cast<unsigned char>(c));
capitalize_next = false;
} else {
c = ::tolower(static_cast<unsigned char>(c));
}
}
return res;
});
}
string string::strip(bool left, bool right, std::optional<const std::string_view> chars) {
static auto strip_part = [](const std::string & s, bool left, bool right, std::optional<const std::string_view> chars) -> std::string {
size_t start = 0;
size_t end = s.length();
auto match_char = [&chars](unsigned char c) -> bool {
return chars ? (*chars).find(c) != std::string::npos : isspace(c);
};
if (left) {
while (start < end && match_char(static_cast<unsigned char>(s[start]))) {
++start;
}
}
if (right) {
while (end > start && match_char(static_cast<unsigned char>(s[end - 1]))) {
--end;
}
}
return s.substr(start, end - start);
};
if (parts.empty()) {
return *this;
}
if (left) {
for (size_t i = 0; i < parts.size(); ++i) {
parts[i].val = strip_part(parts[i].val, true, false, chars);
if (parts[i].val.empty()) {
// remove empty part
parts.erase(parts.begin() + i);
--i;
continue;
} else {
break;
}
}
}
if (right) {
for (size_t i = parts.size(); i-- > 0;) {
parts[i].val = strip_part(parts[i].val, false, true, chars);
if (parts[i].val.empty()) {
// remove empty part
parts.erase(parts.begin() + i);
continue;
} else {
break;
}
}
}
return *this;
}
} // namespace jinja