update run commmand

demo/lac/infer.sh

-python ../../paddlehub/commands/hub.py run hub_module_lac/ --signature lexical_analysis --config resources/test/test.yml  --input_file resources/test/test.csv
+python ../../paddlehub/commands/hub.py run hub_module_lac/ --signature lexical_analysis --config resources/test/test.yml  --input_file resources/test/test.txt

demo/lac/infer_by_code.py

@@ -31,8 +31,8 @@ def infer_with_input_file():
     key = list(data_format.keys())[0]
 
     # parse input file
-    test_csv = os.path.join("resources", "test", "test.csv")
-    test_text = hub.io.reader.csv_reader.read(test_csv)["TEXT_INPUT"]
+    test_file = os.path.join("resources", "test", "test.txt")
+    test_text = hub.io.parser.txt_parser.parse(test_file)
 
     # set input dict
     input_dict = {key: test_text}
@@ -43,4 +43,4 @@ def infer_with_input_file():
 
 
 if __name__ == "__main__":
-    infer_with_input_text()
+    infer_with_input_file()

demo/lac/resources/test/test.csv → demo/lac/resources/test/test.txt

-TEXT_INPUT
 今天是个好日子

demo/senta/infer.sh

-python ../../paddlehub/commands/hub.py run hub_module_senta/ --signature sentiment_classify --config resources/test/test.yml  --dataset resources/test/test.csv
+python ../../paddlehub/commands/hub.py run hub_module_senta/ --signature sentiment_classify --config resources/test/test.yml  --input_file resources/test/test.txt

demo/senta/infer_by_code.py

@@ -31,8 +31,8 @@ def infer_with_input_file():
     key = list(data_format.keys())[0]
 
     # parse input file
-    test_csv = os.path.join("resources", "test", "test.csv")
-    test_text = hub.io.reader.csv_reader.read(test_csv)["TEXT_INPUT"]
+    test_file = os.path.join("resources", "test", "test.txt")
+    test_text = hub.io.parser.txt_parser.parse(test_file)
 
     # set input dict
     input_dict = {key: test_text}

demo/senta/resources/test/test.csv → demo/senta/resources/test/test.txt

-TEXT_INPUT
 这部电影真的很赞
 售后太差！

demo/ssd/infer.sh

-python ../../paddlehub/commands/hub.py run hub_module_ssd/ --signature object_detection --config resources/test/test.yml  --dataset resources/test/test.csv
+python ../../paddlehub/commands/hub.py run hub_module_ssd/ --signature object_detection --config resources/test/test.yml  --input_file resources/test/test.txt

demo/ssd/infer_by_code.py

@@ -30,8 +30,8 @@ def infer_with_input_file():
     key = list(data_format.keys())[0]
 
     # parse input file
-    test_csv = os.path.join("resources", "test", "test.csv")
-    test_images = hub.io.reader.csv_reader.read(test_csv)["IMAGE_PATH"]
+    test_file = os.path.join("resources", "test", "test.txt")
+    test_images = hub.io.parser.txt_parser.parse(test_file)
 
     # set input dict
     input_dict = {key: test_images}

demo/ssd/resources/test/test.csv → demo/ssd/resources/test/test.txt

-IMAGE_PATH
 ./resources/test/test_img_sheep.jpg
 ./resources/test/test_img_cat.jpg
 ./resources/test/test_img_bird.jpg

paddlehub/commands/run.py

@@ -21,7 +21,7 @@ import os
 
 from paddlehub.common.logger import logger
 from paddlehub.commands.base_command import BaseCommand, ENTRY
-from paddlehub.io.reader import csv_reader, yaml_reader
+from paddlehub.io.parser import yaml_parser, txt_parser
 from paddlehub.module.manager import default_module_manager
 from paddlehub.common import utils
 from paddlehub.common.arg_helper import add_argument, print_arguments
@@ -112,7 +112,8 @@ class RunCommand(BaseCommand):
             input_data_key = list(expect_data_format.keys())[0]
             origin_data = {input_data_key: [self.args.data]}
         elif self.args.dataset:
-            origin_data = csv_reader.read(self.args.dataset)
+            input_data_key = list(expect_data_format.keys())[0]
+            origin_data = {input_data_key: txt_parser.parse(self.args.dataset)}
         else:
             print("ERROR! Please specify data to predict.\n")
             print("Summary:\n    %s\n" % module.summary)
@@ -127,7 +128,7 @@ class RunCommand(BaseCommand):
             input_data = {input_data_key: origin_data[origin_data_key]}
             config = {}
         else:
-            yaml_config = yaml_reader.read(self.args.config)
+            yaml_config = yaml_parser.parse(self.args.config)
             if len(expect_data_format) == 1:
                 origin_data_key = list(origin_data.keys())[0]
                 input_data_key = list(expect_data_format.keys())[0]

paddlehub/common/downloader.py

@@ -28,7 +28,6 @@ import tarfile
 
 from paddlehub.common import utils
 from paddlehub.common.logger import logger
-from paddlehub.io.reader import csv_reader
 
 __all__ = ['Downloader']
 FLUSH_INTERVAL = 0.1

paddlehub/common/hub_server.py

@@ -22,7 +22,7 @@ import re
 
 from paddlehub.common import utils
 from paddlehub.common.downloader import default_downloader
-from paddlehub.io.reader import yaml_reader
+from paddlehub.io.parser import yaml_parser
 import paddlehub as hub
 
 RESOURCE_LIST_FILE = "resource_list_file.yml"
@@ -51,7 +51,7 @@ class HubServer:
         if now_time - file_create_time >= CACHE_TIME:
             os.remove(self.resource_list_file_path())
             return False
-        for resource in yaml_reader.read(
+        for resource in yaml_parser.parse(
                 self.resource_list_file_path())['resource_list']:
             for key in resource:
                 if key not in self.resource_list_file:

paddlehub/io/reader.py → paddlehub/io/parser.py

@@ -15,14 +15,14 @@
 import yaml
 
 
-class CSVReader:
+class CSVFileParser:
     def __init__(self):
         pass
 
     def _check(self):
         pass
 
-    def read(self, csv_file):
+    def parse(self, csv_file):
         with open(csv_file, "r") as file:
             content = file.read()
         content = content.split('\n')
@@ -42,18 +42,32 @@ class CSVReader:
         return self.content
 
 
-class YAMLReader:
+class YAMLFileParser:
     def __init__(self):
         pass
 
     def _check(self):
         pass
 
-    def read(self, yaml_file):
+    def parse(self, yaml_file):
         with open(yaml_file, "r") as file:
             content = file.read()
         return yaml.load(content)
 
 
-yaml_reader = YAMLReader()
-csv_reader = CSVReader()
+class TextFileParser:
+    def __init__(self):
+        pass
+
+    def _check(self):
+        pass
+
+    def parse(self, yaml_file):
+        with open(yaml_file, "r") as file:
+            contents = file.read().split("\n")
+        return [content for content in contents if content]
+
+
+csv_parser = CSVFileParser()
+yaml_parser = YAMLFileParser()
+txt_parser = TextFileParser()

paddlehub/module/module.py

@@ -35,7 +35,7 @@ from paddlehub.module.signature import Signature, create_signature
 from paddlehub.module.checker import ModuleChecker
 from paddlehub.module.manager import default_module_manager
 from paddlehub.module.base_processor import BaseProcessor
-from paddlehub.io.reader import yaml_reader
+from paddlehub.io.parser import yaml_parser
 from paddlehub import version
 
 __all__ = ['Module', 'create_module']
@@ -267,7 +267,7 @@ class Module(object):
             if not utils.is_yaml_file(module_info):
                 logger.critical("module info file should in yaml format")
                 exit(1)
-            self.module_info = yaml_reader.read(module_info)
+            self.module_info = yaml_parser.parse(module_info)
         self.author = self.module_info.get('author', 'UNKNOWN')
         self.author_email = self.module_info.get('author_email', 'UNKNOWN')
         self.summary = self.module_info.get('summary', 'UNKNOWN')