Revision 6873
Added by ben leinfelder almost 13 years ago
| src/edu/ucsb/nceas/metacat/dataone/SystemMetadataFactory.java | ||
|---|---|---|
| 29 | 29 |
import java.io.InputStream; |
| 30 | 30 |
import java.math.BigInteger; |
| 31 | 31 |
import java.net.URL; |
| 32 |
import java.net.URLConnection; |
|
| 32 | 33 |
import java.security.NoSuchAlgorithmException; |
| 33 | 34 |
import java.sql.SQLException; |
| 34 | 35 |
import java.util.ArrayList; |
| ... | ... | |
| 357 | 358 |
try {
|
| 358 | 359 |
// download the data from the URL |
| 359 | 360 |
URL dataURL = new URL(dataDocUrl); |
| 360 |
dataObject = dataURL.openStream(); |
|
| 361 |
// TODO: check for valid content |
|
| 361 |
URLConnection dataConnection = dataURL.openConnection(); |
|
| 362 |
|
|
| 363 |
// default is to download the data |
|
| 364 |
dataObject = dataConnection.getInputStream(); |
|
| 365 |
|
|
| 366 |
String detectedContentType = dataConnection.getContentType(); |
|
| 367 |
logMetacat.info("Detected content type: " + detectedContentType);
|
|
| 368 |
|
|
| 369 |
if (detectedContentType != null) {
|
|
| 370 |
// seems to be HTML from the remote location |
|
| 371 |
if (detectedContentType.contains("html")) {
|
|
| 372 |
// if we are not expecting it, we skip it |
|
| 373 |
if (!dataDocMimeType.contains("html")) {
|
|
| 374 |
// set to null so we don't download it |
|
| 375 |
dataObject = null; |
|
| 376 |
logMetacat.warn("Skipping remote resource, unexpected HTML content type at: " + dataDocUrl);
|
|
| 377 |
} |
|
| 378 |
} |
|
| 379 |
|
|
| 380 |
// TODO: any other special processing (csv, images, etc)? |
|
| 381 |
} else {
|
|
| 382 |
// if we don't know what it is, should we skip it? |
|
| 383 |
dataObject = null; |
|
| 384 |
logMetacat.warn("Skipping remote resource, unknown content type at: " + dataDocUrl);
|
|
| 385 |
} |
|
| 386 |
|
|
| 362 | 387 |
} catch (Exception e) {
|
| 363 | 388 |
// error with the download |
| 364 | 389 |
logMetacat.warn("Error downloading remote data. " + e.getMessage());
|
Also available in: Unified diff
do not download and save remote data resources which are HTML but are not expected to be such (login or info/splash pages before data content).
http://bugzilla.ecoinformatics.org/show_bug.cgi?id=5522