如何使python代碼更清潔，優化和高效？

Question

我在python 3中編寫了一個程序，它從圖像中提取元數據，並用這些信息處理多個事情。其中一個特點是它將從元數據中提取GPS座標，並使用谷歌地圖的靜態API來繪製位置地圖。

一切工作正常，但我寫了這個非常混亂的代碼，從找到的數據中提取座標。我如何清理這些代碼使其更有效率？我也很感謝編寫更好代碼的任何提示。

在此先感謝。

這裏的輸入數據的一個例子：（縮短爲節省空間）

EXIF BrightnessValue: 10 
    EXIF ColorSpace: sRGB 
    EXIF ComponentsConfiguration: YCbCr 
    EXIF Contrast: Normal 
    EXIF CustomRendered: Normal 
    EXIF DateTimeDigitized: 2006:10:11 09:37:52 
    EXIF DateTimeOriginal: 2006:10:11 09:37:52 
    EXIF DigitalZoomRatio: 0 
    EXIF ExifImageLength: 768 
    EXIF ExifImageWidth: 1024 
    EXIF ExifVersion: 0221 
    EXIF ExposureBiasValue: 0 
    EXIF ExposureMode: Auto Exposure 
    EXIF ExposureProgram: Aperture Priority 
    EXIF ExposureTime: 1/800 
    EXIF FNumber: 71/10 
    EXIF Flash: Flash did not fire, compulsory flash mode 
    GPS GPSAltitude: 0 
    GPS GPSAltitudeRef: 0 
    GPS GPSLatitude: [33, 51, 2191/100] 
    GPS GPSLatitudeRef: S 
    GPS GPSLongitude: [151, 13, 1173/100] 
    GPS GPSLongitudeRef: E 
    GPS GPSVersionID: [0, 0, 2, 2] 
    Image Artist: 
    Image Copyright: 
    Image DateTime: 2006:10:11 09:37:52 
    Image ExifOffset: 346 
    Image GPSInfo: 946 
    Image ImageDescription: KONICA MINOLTA DIGITAL CAMERA 
    Image Make: Konica Minolta Camera, Inc. 
    Image Model: DiMAGE A2 
    Image Orientation: Horizontal (normal) 

這裏是我的代碼做（輸出）：

['33', '51', '2191/100', 'S', '151', '13', '1173/100', 'E'] 

它提取GPSLatitude ，GPSLatitudeRef，GPSLongitude和GPSLongitudeRef，並按照上述順序將它們分配到單個列表。 （順序很重要，因爲其他功能依賴此列表的順序）

代碼段：

def coordinates_extractor(): 
    out = [] 

    lat1 = [] 
    lon1 = [] 

    lat2 = [] 
    lon2 = [] 

    lat3 = [] 
    lon3 = [] 

    lat4 = [] 
    lon4 = [] 

    ALL = [] 

    for tag in tags.keys(): 
     if tag not in ("JPEGThumbnail", "TIFFThumbnail", "EXIF MakerNote"): 
      out += [("%s: %s" % (tag, tags[tag]))] 
     out = sorted(out) 

    for i in out: 
     if "GPSLatitudeRef:" in i: 
      lat1 += [i] 
     if "GPSLatitude:" in i: 
      lat1 += [i] 

    for i in out: 
     if "GPSLongitudeRef:" in i: 
      lon1 += [i] 
     if "GPSLongitude:" in i: 
      lon1 += [i] 

    for i in lat1: 
     lat2 += i.split() 

    del lat2[0] 
    del lat2[0] 
    del lat2[3] 
    del lat2[3] 

    for i in lat2: 
     if "[" in i: 
      lat3 += [i.replace("[", "")] 
      continue 
     if "]" not in i: 
      lat3 += [i] 
     if "]" in i: 
      lat3 += [i.replace("]", "")] 

    for i in lat3: 
     if "," in i: 
      lat4 += [i.replace(",", "")] 
     if "," not in i: 
      lat4 += [i] 

    for i in lon1: 
     lon2 += i.split() 

    del lon2[0] 
    del lon2[0] 
    del lon2[3] 
    del lon2[3] 

    for i in lon2: 
     if "[" in i: 
      lon3 += [i.replace("[", "")] 
      continue 
     if "]" not in i: 
      lon3 += [i] 
     if "]" in i: 
      lon3 += [i.replace("]", "")] 

    for i in lon3: 
     if "," in i: 
      lon4 += [i.replace(",", "")] 
     if "," not in i: 
      lon4 += [i] 

    LATANDLONG = lat4 + lon4 

    return LATANDLONG 

示例代碼從提取：http://ptforum.photoolsweb.com/ubbthreads.php?ubb=download&Number=1024&filename=1024-2006_1011_093752.jpg

Answer 1

你可以嘗試這樣的事情。我加入了一個嘲弄的字典對象，以便在沒有圖像的情況下輕鬆運行。您只需將其刪除並將其替換爲exifread庫提供的真實tags對象，我相信您正在使用它。

# mock dictionary because I dont have your image 
tags = {} 
tags['GPS GPSLatitude'] = [33, 51, 2191/100] 
tags['GPS GPSLatitudeRef'] = 'S' 
tags['GPS GPSLongitude'] = [151, 13, 1173/100] 
tags['GPS GPSLongitudeRef'] = 'E' 

out = [] 
wantedtags = ('GPS GPSLatitude', 'GPS GPSLatitudeRef', 'GPS GPSLongitude', 'GPS GPSLongitudeRef') 

for tag in wantedtags: 
    try: 
     val = tags[tag] 
     if isinstance(val, list): 
      out.extend(map(str, val)) 
     else: 
      out.append(val) 
    except KeyError: 
     print('Key %s does not exists' % tag) 

print(out) 

儘管一個注意事項是輸出與您的預期輸出完全不同。原因是，在文字2191/100和1173/100在運行時評價爲師語句，然後在列表被視爲21 & 11爲Python 2和21.91 & 11.73爲Python 3.使用from __future__ import division在Python 2中得到更精確的劃分結果。

exifread的行爲有所不同嗎？即它是否以保持小數值的其他形式存儲此值，而不僅僅是一個普通的舊int？

我對自己的模擬字典運行了自己的代碼，輸出和我的版本完全一樣。