添加输出器：csv表格格式

vaqua · Sep 28, 2023 · ec5ae82 · ec5ae82
1 parent a49a78d
commit ec5ae82
Show file tree

Hide file tree

Showing 3 changed files with 56 additions and 1 deletion.
diff --git a/UmiOCR-data/pyapp/ocr/output/output_csv.py b/UmiOCR-data/pyapp/ocr/output/output_csv.py
@@ -0,0 +1,47 @@
+# 输出到csv表格文件
+
+from .output import Output
+
+import csv
+
+
+class OutputCsv(Output):
+    def __init__(self, argd):
+        self.dir = argd["outputDir"]  # 输出路径（文件夹）
+        self.fileName = argd["outputFileName"]  # 文件名
+        self.fileName = self.fileName.replace(
+            "%name", argd["outputDirName"]
+        )  # 文件名添加路径名
+        self.outputPath = f"{self.dir}/{self.fileName}.csv"  # 输出路径
+        self.ingoreBlank = argd["ingoreBlank"]  # 忽略空白文件
+        self.headers = ["Image Name", "OCR", "Image Path"]  # 表头
+        self.lineBreak = "\n"  # 换行符
+        # 创建输出文件
+        try:
+            with open(
+                self.outputPath, "w", encoding="utf-8", newline=""
+            ) as f:  # 覆盖创建文件
+                writer = csv.writer(f)
+                writer.writerow(self.headers)  # 写入CSV表头
+        except Exception as e:
+            raise Exception(f"Failed to create csv file. {e}\n创建csv文件失败。")
+
+    def print(self, res):  # 输出图片结果
+        if not res["code"] == 100 and self.ingoreBlank:
+            return  # 忽略空白图片
+        name = res["fileName"]
+        path = res["path"]
+        textOut = ""
+        if res["code"] == 100:
+            for tb in res["data"]:
+                if tb["text"]:
+                    textOut += tb["text"] + self.lineBreak
+        elif res["code"] == 101:
+            pass
+        else:
+            textOut += f'[Error] OCR failed. Code: {res["code"]}, Msg: {res["data"]}  \n【异常】OCR识别失败。'
+
+        writeList = [name, textOut, path]
+        with open(self.outputPath, "a", encoding="utf-8", newline="") as f:  # 追加写入本地文件
+            writer = csv.writer(f)
+            writer.writerow(writeList)
diff --git a/UmiOCR-data/pyapp/tag_pages/BatchOCR.py b/UmiOCR-data/pyapp/tag_pages/BatchOCR.py
@@ -13,6 +13,7 @@
 from ..ocr.output.output_txt_individual import OutputTxtIndividual
 from ..ocr.output.output_md import OutputMD
 from ..ocr.output.output_jsonl import OutputJsonl
+from ..ocr.output.output_csv import OutputCsv
 
 import os
 import time
@@ -120,6 +121,8 @@ def __initOutputList(self, argd):  # 初始化输出器列表，无异常返回T
                 self.outputList.append(OutputMD(outputArgd))
             if argd["mission.filesType.jsonl"]:  # jsonl
                 self.outputList.append(OutputJsonl(outputArgd))
+            if argd["mission.filesType.csv"]:  # csv
+                self.outputList.append(OutputCsv(outputArgd))
         except Exception as e:
             self.__onEnd(
                 None,

diff --git a/UmiOCR-data/res/qml/TabPages/BatchOCR/BatchOCRConfigs.qml b/UmiOCR-data/res/qml/TabPages/BatchOCR/BatchOCRConfigs.qml
@@ -44,7 +44,7 @@ Configs {
             "datetimeFormat": {
                 "title": qsTr("日期时间格式"),
                 "toolTip": qsTr("文件名中 %date 的日期格式。支持插入以下占位符：\n%Y 年、 %m 月、 %d 日、 %H 小时、 \n%M 分钟、 %S 秒 、 %unix 时间戳 \n举例：%Y年%m月%d日_%H-%M\n生成：2023年09月01日_12-13.txt"),
-                "default": "%Y-%m-%d_%H-%M",
+                "default": "%Y%m%d_%H%M",
                 "advanced": true, // 高级选项
             },
 
@@ -74,6 +74,11 @@ Configs {
                     "toolTip": qsTr("Markdown图文混排格式，可用Markdown阅读器浏览文件"),
                     "default": false,
                 },
+                "csv": {
+                    "title": qsTr("csv 表格文件(Excel)"),
+                    "toolTip": qsTr("将图片信息和识别内容写入csv表格文件。可用Excel打开，另存为xlsx格式。"),
+                    "default": false,
+                },
                 "jsonl": {
                     "title": qsTr("jsonl 原始信息"),
                     "toolTip": qsTr("每行为一条json数据，便于第三方程序读取操作"),