我正在運(yùn)行一個(gè) python CGI 腳本，使用 numpy 從動(dòng)態(tài)創(chuàng)建的 CSV 中讀取數(shù)據(jù)。Numpy 抱怨它找不到文件，但它在那里。從 python 腳本我可以使用 system(cat) 預(yù)先打印文件......filename is: /tmp/file3aQJRnDump file with system(cat)...A-257-1-13629567-97-8867-27041-12323-17473-659Ask numpy to read it...Traceback (most recent call last):  (some lines deleted, then...)  File "request", line 195, in get_rawcsv_column    col_data = get_csv_column(fname,column)  File "request", line 189, in get_csv_column    all_data = np.genfromtxt(fname,delimiter=",")  File "usr/lib/python3.6/site-packages/numpy/lib/npyio.py", line 1551, in genfromtxt  File "usr/lib/python3.6/site-packages/numpy/lib/_datasource.py", line 151, in open  File "usr/lib/python3.6/site-packages/numpy/lib/_datasource.py", line 501, in openOSError: "/tmp/file3aQJRn" not found.之后，我可以在 bash 提示符下轉(zhuǎn)儲(chǔ)文件：# cat /tmp/file3aQJRnA-257-1-13629567-97-8867-27041-12323-17473-659我可以調(diào)用 Python (3.6) 和 numpy 來(lái)讀取文件...numpy.genfromtxt("/tmp/file3aQJRn",delimiter=",")array([             nan,  -2.57000000e+02,  -1.00000000e+00,        -1.36000000e+02,   2.95670000e+04,  -9.70000000e+01,        -8.86700000e+03,  -2.70410000e+04,  -1.23230000e+04,        -1.74730000e+04,  -6.59000000e+02])（ nan 不是問(wèn)題。它是隨后從數(shù)組中刪除的列標(biāo)題。）這里發(fā)生了什么？臨時(shí)文件由 C++ 應(yīng)用程序使用 std::ofstream 從 tmpnam() 獲得臨時(shí)文件名生成。