Skip to content
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.

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

Fix the build breaks the release pipeline and some C++ warnings #372

Merged
merged 8 commits into from
Feb 28, 2023
Merged
Show file tree
Hide file tree
Changes from 6 commits
Commits
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
2 changes: 1 addition & 1 deletion .gitignore
Original file line number Diff line number Diff line change
Expand Up @@ -38,7 +38,7 @@ out/
.scb/
onnxruntime_extensions/_version.py
onnxruntime-*-*-*/
temp_*.onnx
temp_*onnx*
# Java specific ignores
*/.gradle
java/hs_*.log
Expand Down
16 changes: 16 additions & 0 deletions .pipelines/ci.yml
Original file line number Diff line number Diff line change
Expand Up @@ -478,3 +478,19 @@ jobs:
echo "Exception propogation was not enabled correctly."
exit 1
fi


##############################
# Linux for selected_ops build
##############################
- job: Linux_SelectedOpsBuild
pool:
vmImage: 'ubuntu-latest'

steps:
# compiled as only one operator selected.
- bash: |
set -e -x -u
echo 'set (OCOS_ENABLE_BERT_TOKENIZER ON CACHE BOOL "" FORCE)' > cmake/_selectedoplist.cmake
./build.sh -DOCOS_ENABLE_CPP_EXCEPTIONS=OFF -DOCOS_ENABLE_SELECTED_OPLIST=ON
displayName: Build ort-extensions with only one operator was selected
7 changes: 6 additions & 1 deletion CMakeLists.txt
Original file line number Diff line number Diff line change
Expand Up @@ -208,6 +208,9 @@ endif()

if(NOT OCOS_ENABLE_CPP_EXCEPTIONS)
add_compile_definitions(OCOS_NO_EXCEPTIONS ORT_NO_EXCEPTIONS)
if (NOT _ONNXRUNTIME_EMBEDDED)
add_compile_definitions(_HAS_EXCEPTIONS=0)
endif()
endif()

include(FetchContent)
Expand Down Expand Up @@ -254,7 +257,7 @@ if(OCOS_ENABLE_RE2_REGEX)
endif()

# ### scan all source files
set(TARGET_SRC_NOEXCEPTION)
file(GLOB TARGET_SRC_NOEXCEPTION "base/*.h" "base/*.cc")
file(GLOB TARGET_SRC "operators/*.cc" "operators/*.h" "includes/*.h*")

if(OCOS_ENABLE_TF_STRING)
Expand Down Expand Up @@ -402,11 +405,13 @@ standardize_output_folder(ocos_operators)
target_include_directories(noexcep_operators PUBLIC
${ONNXRUNTIME_INCLUDE_DIR}
${PROJECT_SOURCE_DIR}/includes
${PROJECT_SOURCE_DIR}/base
${PROJECT_SOURCE_DIR}/operators)

target_include_directories(ocos_operators PUBLIC
${ONNXRUNTIME_INCLUDE_DIR}
${PROJECT_SOURCE_DIR}/includes
${PROJECT_SOURCE_DIR}/base
${PROJECT_SOURCE_DIR}/operators
${PROJECT_SOURCE_DIR}/operators/tokenizer)

Expand Down
File renamed without changes.
File renamed without changes.
File renamed without changes.
File renamed without changes.
2 changes: 1 addition & 1 deletion operators/string_tensor.cc → base/string_tensor.cc
Original file line number Diff line number Diff line change
@@ -1,8 +1,8 @@
// Copyright (c) Microsoft Corporation. All rights reserved.
// Licensed under the MIT License.
#include "string_tensor.h"
#include "string_utils.h"
#include "ustring.h"
#include "string_tensor.h"

void GetTensorMutableDataString(const OrtApi& api, OrtW::CustomOpApi& ort, OrtKernelContext* context,
const OrtValue* value, std::vector<std::string>& output) {
Expand Down
2 changes: 1 addition & 1 deletion operators/string_tensor.h → base/string_tensor.h
Original file line number Diff line number Diff line change
Expand Up @@ -3,8 +3,8 @@

#pragma once

#include <string>
#include "ocos.h"
#include <string>


// Retrieves a vector of strings if the input type is std::string.
Expand Down
File renamed without changes.
File renamed without changes.
3 changes: 2 additions & 1 deletion operators/ustring.cc → base/ustring.cc
Original file line number Diff line number Diff line change
@@ -1,7 +1,8 @@
// Copyright (c) Microsoft Corporation. All rights reserved.
// Licensed under the MIT License.
#include <iostream>
#include "ustring.h"
#include <iostream>


ustring::ustring() : std::u32string() {
}
Expand Down
9 changes: 3 additions & 6 deletions operators/ustring.h → base/ustring.h
Original file line number Diff line number Diff line change
@@ -1,15 +1,12 @@
// Copyright (c) Microsoft Corporation. All rights reserved.
// Licensed under the MIT License.
#pragma once
#include <string>
#include <locale>
#include <functional>

#define _SILENCE_CXX17_CODECVT_HEADER_DEPRECATION_WARNING 1
#include <codecvt>

#include "ocos.h"
#include <locale>
#include <codecvt>

// ustring needs a new implementation, due to the std::codecvt deprecation.
// Wrap u32string with ustring, in case we will use other implementation in the future
class ustring : public std::u32string {
public:
Expand Down
2 changes: 2 additions & 0 deletions includes/ocos.h
Original file line number Diff line number Diff line change
Expand Up @@ -3,6 +3,8 @@

#pragma once

#define _SILENCE_CXX17_CODECVT_HEADER_DEPRECATION_WARNING
#include <string>
#include <algorithm>
#include <functional>
#include <iterator>
Expand Down
3 changes: 1 addition & 2 deletions operators/tokenizer/basic_tokenizer.cc
Original file line number Diff line number Diff line change
@@ -1,12 +1,11 @@
// Copyright (c) Microsoft Corporation. All rights reserved.
// Licensed under the MIT License.

#include "string_utils.h"
#include "basic_tokenizer.hpp"
#include "string_utils.h"
#include "string_tensor.h"
#include <vector>
#include <locale>
#include <codecvt>
#include <algorithm>

BasicTokenizer::BasicTokenizer(bool do_lower_case, bool tokenize_chinese_chars, bool strip_accents,
Expand Down
2 changes: 0 additions & 2 deletions operators/tokenizer/bert_tokenizer.hpp
Original file line number Diff line number Diff line change
Expand Up @@ -3,8 +3,6 @@

#pragma once

#include <unordered_map>
#include <vector>
#include "ocos.h"
#include "ustring.h"
#include "string_utils.h"
Expand Down
2 changes: 0 additions & 2 deletions operators/tokenizer/bert_tokenizer_decoder.hpp
Original file line number Diff line number Diff line change
Expand Up @@ -3,8 +3,6 @@

#pragma once

#include <unordered_map>
#include <vector>
#include "ocos.h"
#include "ustring.h"
#include "string_utils.h"
Expand Down
65 changes: 47 additions & 18 deletions operators/tokenizer/bpetokenizer.hpp
Original file line number Diff line number Diff line change
@@ -1,28 +1,58 @@
// Licensed under the MIT License.
// Partial code comes from other Microsoft employee.
#pragma once
#include <string>
#include <vector>
#include <fstream>
#include <sstream>
#include <iostream>
#include <algorithm>
#include <list>
#include <memory>
#include "ocos.h"
#include "ustring.h"

#include <regex>
#include <sstream>
#include <stdexcept>
#include <list>
Copy link
Contributor

Choose a reason for hiding this comment

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

Perfect - was going to say this is what caused the ci failures in the previous run, should be fine now.

#include <unordered_map>
#include <functional>
#include <codecvt>
#include <mutex>

#include "unicode.h"
#include "nlohmann/json.hpp"
#include "clip_tokenizer.hpp"
#include "gpt2_tokenizer.hpp"
#include "roberta_tokenizer.hpp"
#include "string_utils.h"
#include "string_tensor.h"
#include "unicode.h"

// Note: the following logic comes from CPython: unicodetype_db.h (_PyUnicode_IsWhitespace)
inline bool IsUnicodeSpace(char32_t ch) {
switch (ch) {
case 0x0009:
case 0x000A:
case 0x000B:
case 0x000C:
case 0x000D:
case 0x001C:
case 0x001D:
case 0x001E:
case 0x001F:
case 0x0020:
case 0x0085:
case 0x00A0:
case 0x1680:
case 0x2000:
case 0x2001:
case 0x2002:
case 0x2003:
case 0x2004:
case 0x2005:
case 0x2006:
case 0x2007:
case 0x2008:
case 0x2009:
case 0x200A:
case 0x2028:
case 0x2029:
case 0x202F:
case 0x205F:
case 0x3000:
return true;
}
return false;
}

inline bool IsEmptyUString(const ustring& str) {
return std::all_of(str.begin(), str.end(), [](char32_t ch) { return IsUnicodeSpace(ch); });
}

class SpecialTokenMap {
public:
Expand Down Expand Up @@ -117,7 +147,6 @@ class VocabData {
} else {
int id = static_cast<int>(vocab_map_.size());
vocab_map_[unk_token] = id;
std::cerr << "Special token (" << unk_token << ") have been added in the vocabulary." << std::endl;
}

std::wstring_convert<std::codecvt_utf8<char32_t>, char32_t> str_convert;
Expand Down
66 changes: 3 additions & 63 deletions operators/tokenizer/clip_tokenizer.cc
Original file line number Diff line number Diff line change
@@ -1,68 +1,8 @@
// Copyright (c) Microsoft Corporation. All rights reserved.
// Licensed under the MIT License.
// Partial code comes from other Microsoft employee.

#include <string>
#include <vector>
#include <fstream>
#include <sstream>
#include <iostream>
#include <algorithm>
#include <list>
#include <memory>
#include <regex>
#include <sstream>
#include <stdexcept>
#include <unordered_map>
#include <functional>
#include <codecvt>
#include <mutex>

#include "nlohmann/json.hpp"
#include "bpetokenizer.hpp"
#include "string_tensor.h"
#include "unicode.h"

// Note: the following logic comes from CPython: unicodetype_db.h (_PyUnicode_IsWhitespace)
bool IsInUnicodeSpace(char32_t ch) {
switch (ch) {
case 0x0009:
case 0x000A:
case 0x000B:
case 0x000C:
case 0x000D:
case 0x001C:
case 0x001D:
case 0x001E:
case 0x001F:
case 0x0020:
case 0x0085:
case 0x00A0:
case 0x1680:
case 0x2000:
case 0x2001:
case 0x2002:
case 0x2003:
case 0x2004:
case 0x2005:
case 0x2006:
case 0x2007:
case 0x2008:
case 0x2009:
case 0x200A:
case 0x2028:
case 0x2029:
case 0x202F:
case 0x205F:
case 0x3000:
return true;
}
return false;
}

bool IsEmptyUstring(const ustring& str) {
return std::all_of(str.begin(), str.end(), [](char32_t ch) { return IsInUnicodeSpace(ch); });
}
#include "clip_tokenizer.hpp"
#include "string_utils.h"

KernelClipBpeTokenizer::KernelClipBpeTokenizer(const OrtApi& api, const OrtKernelInfo& info)
: BaseKernel(api, info) {
Expand Down Expand Up @@ -93,7 +33,7 @@ KernelClipBpeTokenizer::KernelClipBpeTokenizer(const OrtApi& api, const OrtKerne
std::vector<int64_t> KernelClipBpeTokenizer::Tokenize(ustring& input, int64_t max_length) {
std::vector<int64_t> res;

if (IsEmptyUstring(input)) {
if (IsEmptyUString(input)) {
return res;
}
// Add <|startoftext|> token to result
Expand Down
9 changes: 4 additions & 5 deletions operators/tokenizer/clip_tokenizer.hpp
Original file line number Diff line number Diff line change
@@ -1,9 +1,8 @@
#include <list>
#include "ocos.h"
#include "ustring.h"
#include "string_utils.h"
// Copyright (c) Microsoft Corporation. All rights reserved.
// Licensed under the MIT License.

class VocabData;
#pragma once
#include "bpetokenizer.hpp"

struct KernelClipBpeTokenizer : BaseKernel {
KernelClipBpeTokenizer(const OrtApi& api, const OrtKernelInfo& info);
Expand Down
Loading