wenbingl
commented
8 months ago what's the compiler that you are using, it looks uint8_t type definition is missed, which was added in C++ 11 standard. You may try to explicitly add
Closed aejazm closed 5 months ago
wenbingl
commented
8 months ago what's the compiler that you are using, it looks uint8_t type definition is missed, which was added in C++ 11 standard. You may try to explicitly add
KilowattJunkie
commented
8 months ago I also ran into this issue when trying to run the library build. I'm running on a basic install of Arch Linux, using GCC 13, on the current HEAD (d47a3dd)
./build_lib.sh --config Release --update --build --parallel/usr/bin/c++ --version
c++ (GCC) 13.2.1 20230801I had to add <cstdint> to base64.h, and remove the std:: prefix from some of the types in segment_extraction.cc (which makes them consistent with other usages in the file), to get it compiled:
diff --git a/base/base64.h b/base/base64.h
index 121b223..e29290f 100644
--- a/base/base64.h
+++ b/base/base64.h
@@ -4,6 +4,7 @@
#pragma once
#include <string>
#include <vector>
+#include <cstdint>
bool base64_encode(const std::vector<uint8_t>& input, std::string& encoded);
bool base64_decode(const std::string& encoded, std::vector<uint8_t>& raw);
diff --git a/operators/math/segment_extraction.cc b/operators/math/segment_extraction.cc
index 1d4871a..3d880b6 100644
--- a/operators/math/segment_extraction.cc
+++ b/operators/math/segment_extraction.cc
@@ -11,9 +11,9 @@ OrtStatusPtr segment_extraction(const ortc::Tensor<int64_t>& input,
return OrtW::CreateStatus("[SegmentExtraction]: Expect input dimension [n] or [1,n].", ORT_INVALID_GRAPH);
}
const int64_t* p_data = input.Data();
- std::vector<std::int64_t> segment_value;
- std::vector<std::int64_t> segment_position;
- for (std::int64_t i = 0; i < input.NumberOfElement(); i++) {
+ std::vector<int64_t> segment_value;
+ std::vector<int64_t> segment_position;
+ for (int64_t i = 0; i < input.NumberOfElement(); i++) {
if (!p_data[i]) {
continue;
}I'm not going to make a PR only because I'm not sure how the changes will affect things (especially on other platforms), but hopefully this helps if anyone else sees the same error.
leejuyuu
commented
6 months ago I am also having the same build error on Arch and GCC 13.2.1.
wenbingl
commented
6 months ago @aejazm, @KilowattJunkie, @leejuyuu , I don't have a machine with GCC-13 installed, can any of you help to validate the PR I linked here?
leejuyuu
commented
6 months ago @wenbingl I can build this branch successfully on Arch Linux, with GCC version as follows.
➜ gcc -v
Using built-in specs.
COLLECT_GCC=gcc
COLLECT_LTO_WRAPPER=/usr/lib/gcc/x86_64-pc-linux-gnu/13.2.1/lto-wrapper
Target: x86_64-pc-linux-gnu
Configured with: /build/gcc/src/gcc/configure --enable-languages=ada,c,c++,d,fortran,go,lto,m2,objc,obj-c++ --enable-bootstrap --prefix=/usr --libdir=/usr/lib --libexecdir=/usr/lib --mandir=/usr/share/man --infodir=/usr/share/info --with-bugurl=https://bugs.archlinux.org/ --with-build-config=bootstrap-lto --with-linker-hash-style=gnu --with-system-zlib --enable-__cxa_atexit --enable-cet=auto --enable-checking=release --enable-clocale=gnu --enable-default-pie --enable-default-ssp --enable-gnu-indirect-function --enable-gnu-unique-object --enable-libstdcxx-backtrace --enable-link-serialization=1 --enable-linker-build-id --enable-lto --enable-multilib --enable-plugin --enable-shared --enable-threads=posix --disable-libssp --disable-libstdcxx-pch --disable-werror
Thread model: posix
Supported LTO compression algorithms: zlib zstd
gcc version 13.2.1 20230801 (GCC) The tests passed, but I see some warnings
wenbingl
commented
6 months ago Thanks for verification. The warnings are irrelevant which are caused by the glitches in the test onnx models.
leejuyuu
commented
6 months ago Got it. Thanks for fixing this!
Getting the following errors while building the HEAD version on Linux following instructions from here. I am attaching the full build output as well. Please let me know if I am missing something basic.
$ git clone https://github.com/microsoft/onnxruntime-extensions $ cd onnxruntime-extensions $ ./build_lib.sh --config Release --update --build --parallel .................. [ 42%] Building CXX object _deps/spm-build/src/CMakeFiles/sentencepiece-static.dir/normalizer.cc.o [ 42%] Building CXX object _deps/opencv-build/modules/core/CMakeFiles/opencv_core.dir/src/algorithm.cpp.o In file included from ./sources/onnxruntime-extensions/base/base64.cc:3: ./sources/onnxruntime-extensions/base/base64.h:8:38: error: ‘uint8_t’ was not declared in this scope 8 | bool base64_encode(const std::vector& input, std::string& encoded);
| ^’; did you forget to ‘#include ’?
6 | #include
+++ |+#include
7 |
./sources/onnxruntime-extensions/base/base64.h:8:45: error: template argument 1 is invalid
8 | bool base64_encode(const std::vector& input, std::string& encoded);
| ^
./sources/onnxruntime-extensions/base/base64.h:8:45: error: template argument 2 is invalid
./sources/onnxruntime-extensions/base/base64.h:9:60: error: ‘uint8_t’ was not declared in this scope
9 | bool base64_decode(const std::string& encoded, std::vector& raw);
| ^’; did you forget to ‘#include ’?
./sources/onnxruntime-extensions/base/base64.h:9:67: error: template argument 1 is invalid
9 | bool base64_decode(const std::string& encoded, std::vector& raw);
| ^
./sources/onnxruntime-extensions/base/base64.h:9:67: error: template argument 2 is invalid
./sources/onnxruntime-extensions/base/base64.cc:9:38: error: ‘uint8_t’ was not declared in this scope
9 | bool base64_encode(const std::vector& input, std::string& encoded) {
| ^’; did you forget to ‘#include ’?
4 | #include
+++ |+#include
5 |
./sources/onnxruntime-extensions/base/base64.cc:9:45: error: template argument 1 is invalid
9 | bool base64_encode(const std::vector& input, std::string& encoded) {
| ^
./sources/onnxruntime-extensions/base/base64.cc:9:45: error: template argument 2 is invalid
./sources/onnxruntime-extensions/base/base64.cc: In function ‘bool base64_encode(const int&, std::string&)’:
./sources/onnxruntime-extensions/base/base64.cc:11:27: error: request for member ‘size’ in ‘input’, which is of non-class type ‘const int’
11 | encoded.reserve(((input.size() / 3) + (input.size() % 3 > 0)) 4);
| ^~~~
./sources/onnxruntime-extensions/base/base64.cc:11:48: error: request for member ‘size’ in ‘input’, which is of non-class type ‘const int’
11 | encoded.reserve(((input.size() / 3) + (input.size() % 3 > 0)) 4);
| ^~~~
./sources/onnxruntime-extensions/base/base64.cc:12:3: error: ‘uint32_t’ was not declared in this scope
12 | uint32_t temp;
| ^’; did you forget to ‘#include ’?
./sources/onnxruntime-extensions/base/base64.cc:13:15: error: ‘uint8_t’ was not declared in this scope
13 | std::vector::const_iterator cursor = input.begin();
| ^’; did you forget to ‘#include ’?
./sources/onnxruntime-extensions/base/base64.cc:13:22: error: template argument 1 is invalid
13 | std::vector::const_iterator cursor = input.begin();
| ^
./sources/onnxruntime-extensions/base/base64.cc:13:22: error: template argument 2 is invalid
./sources/onnxruntime-extensions/base/base64.cc:13:40: error: expected initializer before ‘cursor’
13 | std::vector::const_iterator cursor = input.begin();
| ^& decoded) {
| ^’; did you forget to ‘#include ’?
./sources/onnxruntime-extensions/base/base64.cc:43:65: error: template argument 1 is invalid
43 | bool base64_decode(const std::string& input, std::vector& decoded) {
| ^
./sources/onnxruntime-extensions/base/base64.cc:43:65: error: template argument 2 is invalid
./sources/onnxruntime-extensions/base/base64.cc: In function ‘bool base64_decode(const std::string&, int&)’:
./sources/onnxruntime-extensions/base/base64.cc:54:11: error: request for member ‘clear’ in ‘decoded’, which is of non-class type ‘int’
54 | decoded.clear();
| ^~~~~
./sources/onnxruntime-extensions/base/base64.cc:55:11: error: request for member ‘reserve’ in ‘decoded’, which is of non-class type ‘int’
55 | decoded.reserve(((input.length() / 4) * 3) - padding);
| ^’; did you forget to ‘#include ’?
./sources/onnxruntime-extensions/base/base64.cc:61:7: error: ‘temp’ was not declared in this scope; did you mean ‘tm’?
61 | temp <<= 6;
| ^~~~
| tm
./sources/onnxruntime-extensions/base/base64.cc:75:21: error: request for member ‘push_back’ in ‘decoded’, which is of non-class type ‘int’
75 | decoded.push_back((temp >> 16) & 0x000000FF);
| ^
~~ ./sources/onnxruntime-extensions/base/base64.h:7:1: note: ‘uint8_t’ is defined in header ‘~~ ./sources/onnxruntime-extensions/base/base64.h:9:60: note: ‘uint8_t’ is defined in header ‘~~ ./sources/onnxruntime-extensions/base/base64.cc:5:1: note: ‘uint8_t’ is defined in header ‘~~~ ./sources/onnxruntime-extensions/base/base64.cc:12:3: note: ‘uint32_t’ is defined in header ‘~~ ./sources/onnxruntime-extensions/base/base64.cc:13:15: note: ‘uint8_t’ is defined in header ‘~./sources/onnxruntime-extensions/base/base64.cc:14:36: error: request for member ‘size’ in ‘input’, which is of non-class type ‘const int’ 14 | for (size_t idx = 0; idx < input.size() / 3; idx++) { | ^~~~ ./sources/onnxruntime-extensions/base/base64.cc:15:5: error: ‘temp’ was not declared in this scope; did you mean ‘tm’? 15 | temp = (cursor++) << 16; //Convert to big endian | ^~~~ | tm ./sources/onnxruntime-extensions/base/base64.cc:15:14: error: ‘cursor’ was not declared in this scope 15 | temp = (cursor++) << 16; //Convert to big endian | ^~./sources/onnxruntime-extensions/base/base64.cc:23:17: error: request for member ‘size’ in ‘input’, which is of non-class type ‘const int’ 23 | switch (input.size() % 3) { | ^~~~ ./sources/onnxruntime-extensions/base/base64.cc:25:7: error: ‘temp’ was not declared in this scope; did you mean ‘tm’? 25 | temp = (cursor++) << 16; | ^~~~ | tm ./sources/onnxruntime-extensions/base/base64.cc:25:16: error: ‘cursor’ was not declared in this scope 25 | temp = (cursor++) << 16; | ^~./sources/onnxruntime-extensions/base/base64.cc: At global scope: ./sources/onnxruntime-extensions/base/base64.cc:43:58: error: ‘uint8_t’ was not declared in this scope 43 | bool base64_decode(const std::string& input, std::vector~~ ./sources/onnxruntime-extensions/base/base64.cc:43:58: note: ‘uint8_t’ is defined in header ‘~~ ./sources/onnxruntime-extensions/base/base64.cc:56:3: error: ‘uint32_t’ was not declared in this scope 56 | uint32_t temp = 0; | ^~~~ ./sources/onnxruntime-extensions/base/base64.cc:56:3: note: ‘uint32_t’ is defined in header ‘~~~~ ./sources/onnxruntime-extensions/base/base64.cc:76:21: error: request for member ‘push_back’ in ‘decoded’, which is of non-class type ‘int’ 76 | decoded.push_back((temp >> 8) & 0x000000FF); | ^~~~~ ./sources/onnxruntime-extensions/base/base64.cc:79:21: error: request for member ‘push_back’ in ‘decoded’, which is of non-class type ‘int’ 79 | decoded.push_back((temp >> 10) & 0x000000FF); | ^~~~~ ./sources/onnxruntime-extensions/base/base64.cc:88:13: error: request for member ‘push_back’ in ‘decoded’, which is of non-class type ‘int’ 88 | decoded.push_back((temp >> 16) & 0x000000FF); | ^~~~~ ./sources/onnxruntime-extensions/base/base64.cc:88:24: error: ‘temp’ was not declared in this scope; did you mean ‘tm’? 88 | decoded.push_back((temp >> 16) & 0x000000FF); | ^~~~ | tm ./sources/onnxruntime-extensions/base/base64.cc:89:13: error: request for member ‘push_back’ in ‘decoded’, which is of non-class type ‘int’ 89 | decoded.push_back((temp >> 8) & 0x000000FF); | ^~~~~ ./sources/onnxruntime-extensions/base/base64.cc:90:13: error: request for member ‘push_back’ in ‘decoded’, which is of non-class type ‘int’ 90 | decoded.push_back((temp)&0x000000FF); | ^~~~~ buildOutput.txt