From 8354f1a7a2798f0154d8a8946040baf0f6f649c4 Mon Sep 17 00:00:00 2001
From: Mathieu Loiseau <mathieu.loiseau@liris.cnrs.fr>
Date: Tue, 18 Jun 2024 14:33:31 +0200
Subject: [PATCH] =?UTF-8?q?temps=20d'ex=C3=A9=20optionnels=20dans=20wrappe?=
=?UTF-8?q?r?=
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit
---
README.md | 22 ++++++++++++++++++++
src/wiktextract_wrapper.py | 41 ++++++++++++++++++++++++++++++--------
2 files changed, 55 insertions(+), 8 deletions(-)
diff --git a/README.md b/README.md
index cb07af7..1288722 100755
--- a/README.md
+++ b/README.md
@@ -46,10 +46,12 @@ pip install -r requirements.txt
_Since `wiktextract` and its dependency `wikitextprocessor` are not regularly published as a Python package, it's a challenge to fix them to a specific version. From `requirements.txt`, the latest version will always be installed. Attention: This might mean that after reinstalling, the output schema of `wiktextract` might have slightly changed._
### 4. Congigure server
+
[config.py](https://gitlab.liris.cnrs.fr/lex-game/live-query-wiktextract/-/blob/main/src/config.py) contains :
* server settings (`host`, `port` and `debug` (boolean))
* supported wiktionary language
* working directory (this can be useful if the server is launched by another server using absolute paths to handle virtual environment)
+
### 5. Load templates from dump files
Run the script `src/load_dumps.py` to load the most recent dumpfile (for each [supported wiktionary language](https://gitlab.liris.cnrs.fr/lex-game/live-query-wiktextract/-/blob/main/src/config.py#L5)) into an sqlite database that will be used by `wiktextract`.
@@ -73,6 +75,26 @@ You can run directly in your virtual environment using absolute paths (in case a
sh -c nohup /var/www/live-query-wiktextract/lq-w-extr/bin/python3 /var/www/live-query-wiktextract/src/app.py
```
+## Use without server
+usage: wiktextract_wrapper.py [-h] [-l WORD_LANGUAGE] [-w WIKT_LANGUAGE]
+ [-e ENTRY] [-z] [-A]
+```
+Wiktextract wrapper
+ ex :
+ ‣python3 src/wiktextract_wrapper.py -l en -w fr -e yellow
+
+options:
+ -h, --help show this help message and exit
+ -l WORD_LANGUAGE, --word_language WORD_LANGUAGE
+ language of the sought entry
+ -w WIKT_LANGUAGE, --wikt_language WIKT_LANGUAGE
+ language of the wiktionary
+ -e ENTRY, --entry ENTRY
+ the entry
+ -z, --zero_config Don't use if you know how to configure a server (this changes the working directory)
+ -A, --force_ascii json avec que des caractères ascii
+ -t, --show_timings montrer les temps d'exécution
+```
## Using Docker
diff --git a/src/wiktextract_wrapper.py b/src/wiktextract_wrapper.py
index c458dc4..2d6bc48 100755
--- a/src/wiktextract_wrapper.py
+++ b/src/wiktextract_wrapper.py
@@ -10,6 +10,8 @@ if __name__ == "__main__":
parser.add_argument("-e", "--entry", help="the entry", type=str, default=None)
parser.add_argument("-z", "--zero_config", help="Don't use if you know how to configure a server (this changes the working directory)", action="store_true")
parser.add_argument("-A", "--force_ascii", help="json avec que des caractères ascii", action="store_true")
+ parser.add_argument("-t", "--show_timings", help="montrer les temps d'exécution", action="store_true")
+
args = parser.parse_args()
formerdir = None
if args.zero_config:
@@ -19,12 +21,20 @@ if __name__ == "__main__":
formerdir = os.getcwd()
os.chdir(pathlib.Path(__file__).parent.parent.resolve())
-
+import time
+t = time.time()
from wiktextract.page import parse_page
+if args.show_timings:
+ print(time.time()-t)
from wiktextract_context import get_wiktextract_context
import json
from importlib import metadata
import git
+wiktextractime = -1
+convertime = -1
+startime = -1
+startparsetime = -1
+getpagetime = -1
class Wiktextract:
def __init__(self, wiktlang: str, wordlang: str):
@@ -33,18 +43,26 @@ class Wiktextract:
self.wxr = get_wiktextract_context(wiktlang, wordlang)
- def parse_page(self, title: str, wikstraktor_format: bool = True):
+ def parse_page(self, title: str):
+ global wiktextractime
+ global convertime
+ global startime
+ global startparsetime
+ global getpagetime
+ startparsetime = time.time()
page = self.wxr.wtp.get_page(title)
if not page:
return None
-
-
+ getpagetime = time.time()
result = parse_page(self.wxr, title, page.body)
- if wikstraktor_format and result:
- result = self.wikstraktor_format(result)
- return result
+ wiktextractime = time.time()
+ converted_result = self.convert(result)
+ convertime = time.time()
+
+ return converted_result
+ #return result
- def wikstraktor_format(self, data_format1):
+ def convert(self, data_format1):
transformed_data = []
for index, pos in enumerate(data_format1):
@@ -108,10 +126,17 @@ class Wiktextract:
if __name__ == "__main__":
+ top = time.time()
if args.entry != None:
+ startime = time.time()
wkstrkt = Wiktextract(args.wikt_language, args.word_language)
+ instantiatime = time.time()
result = wkstrkt.parse_page(args.entry)
print(json.dumps(result, ensure_ascii=args.force_ascii))
+ endtime = time.time()
+ if args.show_timings:
+ print(f"Execution ({endtime-startime}â€):\n\tinstantiation→{instantiatime-startime}\n\twiktextract get page → {getpagetime-startparsetime}\n\twiktextract parse → {wiktextractime-startparsetime}\n\tconversion → {convertime-wiktextractime}\n\tprint result → {endtime - convertime}")
+ print(time.time()-top)
else:
print("{'err':'You need to specify a word'}")
if formerdir != None:
--
GitLab