1 | package org.greenstone.atea;
|
---|
2 |
|
---|
3 | import java.util.*;
|
---|
4 | import java.io.*;
|
---|
5 |
|
---|
6 | import org.apache.log4j.Logger;
|
---|
7 |
|
---|
8 | import org.greenstone.util.SafeProcess;
|
---|
9 |
|
---|
10 | /**
|
---|
11 | * Runs some of the important mongoDB queries I ran.
|
---|
12 | *
|
---|
13 | * This program expects a folder ../mongo-data-auto to exist.
|
---|
14 | *
|
---|
15 | * TO COMPILE OR RUN, FIRST DO:
|
---|
16 | * cd maori-lang-detection/apache-opennlp-1.9.1
|
---|
17 | * export OPENNLP_HOME=`pwd`
|
---|
18 | * cd maori-lang-detection/src
|
---|
19 | *
|
---|
20 | * TO COMPILE:
|
---|
21 | * maori-lang-detection/src$
|
---|
22 | * javac -cp ".:../conf:../lib/*" org/greenstone/atea/SummaryTool.java
|
---|
23 | *
|
---|
24 | * TO RUN:
|
---|
25 | * maori-lang-detection/src$
|
---|
26 | * java -cp ".:../conf:../lib/*" org/greenstone/atea/SummaryTool
|
---|
27 | * OR:
|
---|
28 | * java -cp ".:../conf:../lib/*" org/greenstone/atea/SummaryTool ../mongodb-data/domainsNZ_IsMRI.txt
|
---|
29 | *
|
---|
30 | */
|
---|
31 | public class SummaryTool {
|
---|
32 |
|
---|
33 | static Logger logger = Logger.getLogger(org.greenstone.atea.SummaryTool.class.getName());
|
---|
34 |
|
---|
35 |
|
---|
36 | static private final String GEOJSON_FEATURES_FILE_PREFIX = "geojson-features_";
|
---|
37 |
|
---|
38 | static private final long FIXED_SEED = 1000;
|
---|
39 |
|
---|
40 | private final MongoDBQueryer mongodbQueryer;
|
---|
41 | private File outFolder;
|
---|
42 |
|
---|
43 |
|
---|
44 | public static class Tuple {
|
---|
45 | public final String url;
|
---|
46 | public final String countryCode;
|
---|
47 |
|
---|
48 | public Tuple(String url, String countryCode) {
|
---|
49 | this.url = url;
|
---|
50 | this.countryCode = countryCode;
|
---|
51 | }
|
---|
52 |
|
---|
53 | public String toString() {
|
---|
54 | return this.url + "," + countryCode;
|
---|
55 | }
|
---|
56 | }
|
---|
57 |
|
---|
58 |
|
---|
59 | public SummaryTool(MongoDBQueryer mongodbQueryer, File outFolder)
|
---|
60 | {
|
---|
61 | this.mongodbQueryer = mongodbQueryer;
|
---|
62 | this.outFolder = outFolder;
|
---|
63 | }
|
---|
64 |
|
---|
65 |
|
---|
66 | public void produceURLsForPagesInMRI(File domainsFile) {
|
---|
67 | ArrayList<Tuple> urlsList = getURLsForAllWebPagesInSiteListing(MongoDBQueryer.IS_MRI, domainsFile);
|
---|
68 | File outFile = new File(outFolder, "isMRI_"+domainsFile.getName());
|
---|
69 | writeURLsToFile(urlsList, outFile, urlsList.size());
|
---|
70 |
|
---|
71 | System.out.println("Wrote all isMRI web page URLs for the sites in input domainsFile\ninto file: "
|
---|
72 | + Utility.getFilePath(outFile));
|
---|
73 | }
|
---|
74 |
|
---|
75 | public void produceURLsForPagesContainingMRI(File domainsFile) {
|
---|
76 | ArrayList<Tuple> urlsList = getURLsForAllWebPagesInSiteListing(MongoDBQueryer.CONTAINS_MRI, domainsFile);
|
---|
77 | File outFile = new File(outFolder, "containsMRI_"+domainsFile.getName());
|
---|
78 | writeURLsToFile(urlsList, outFile, urlsList.size());
|
---|
79 |
|
---|
80 | System.out.println("Wrote all containsMRI web page URLs for the sites in input domainsFile\ninto file: "
|
---|
81 | + Utility.getFilePath(outFile));
|
---|
82 | }
|
---|
83 |
|
---|
84 | private ArrayList<Tuple> getURLsForAllWebPagesInSiteListing(int filterType, File domainsFile) {
|
---|
85 | ArrayList<Tuple> urlsList = new ArrayList<Tuple>();
|
---|
86 |
|
---|
87 | // 1. read each url from the domainsFile
|
---|
88 | // 1a. do the query
|
---|
89 | // 1b. add the arraylist result to urls
|
---|
90 |
|
---|
91 | try (
|
---|
92 | BufferedReader reader = new BufferedReader(new FileReader(domainsFile));
|
---|
93 | ) {
|
---|
94 |
|
---|
95 | String domain;
|
---|
96 |
|
---|
97 | while((domain = reader.readLine()) != null) {
|
---|
98 | domain = domain.trim();
|
---|
99 | if(!domain.equals("")) {
|
---|
100 |
|
---|
101 | String countryCode = "";
|
---|
102 | int index = domain.lastIndexOf(",");
|
---|
103 | if(index != -1) {
|
---|
104 | countryCode = domain.substring(index+1).trim();
|
---|
105 | domain = domain.substring(0, index);
|
---|
106 | }
|
---|
107 | ArrayList<String> moreURLs = mongodbQueryer.queryAllMatchingURLsFilteredBy(domain, filterType);
|
---|
108 |
|
---|
109 | // Print out whether there were no isMRI pages for the domain (only containsMRI). A useful thing to know
|
---|
110 | if(moreURLs.size() == 0 && filterType == MongoDBQueryer.IS_MRI) {
|
---|
111 | System.out.println(" " + countryCode + " domain " + domain + " had no webpages where isMRI=true - only containsMRI.");
|
---|
112 | }
|
---|
113 |
|
---|
114 | //urlsList.addAll(moreURLs);
|
---|
115 | for(int i = 0; i < moreURLs.size(); i++) {
|
---|
116 | urlsList.add(new Tuple(moreURLs.get(i), countryCode));
|
---|
117 | }
|
---|
118 |
|
---|
119 | }
|
---|
120 | }
|
---|
121 | System.err.println("");
|
---|
122 | } catch(Exception e) {
|
---|
123 | logger.error("Unable to read URLs from file " + Utility.getFilePath(domainsFile));
|
---|
124 | logger.error(e.getMessage(), e);
|
---|
125 | }
|
---|
126 |
|
---|
127 | return urlsList;
|
---|
128 | }
|
---|
129 |
|
---|
130 | /** Given a hand curated list of all sites with positive numPagesContainingMRI
|
---|
131 | * determined by manual inspection, get a listing of all their web pages that
|
---|
132 | * are IN_MRI (or CONTAINS_MRI?).
|
---|
133 | * Total all these pages that are inMRI (N), then work out the correct sample size (n)
|
---|
134 | * at 90% confidence with 5% margin of error. Then generate a random listing
|
---|
135 | * of n of these pages in MRI of these trusted sites and output to a file
|
---|
136 | * for manual inspection of the sample webpage URLs at page-level. */
|
---|
137 | /* OLD: Given a hand curated list of non-NZ sites that CONTAINS_MRI, get a listing
|
---|
138 | * of all their web pages IN_MRI (or CONTAINS_MRI).
|
---|
139 | * Plus a listing of all the NZ pages IN_MRI. */
|
---|
140 | //public void webPagesOfAllNZSitesAndDomainListing(File domainsFile) {
|
---|
141 | public void mriWebPageListingForDomainListing(File domainsFile) {
|
---|
142 |
|
---|
143 | int filterType = MongoDBQueryer.IS_MRI;
|
---|
144 |
|
---|
145 | // for overseas websites,
|
---|
146 | //produceURLsForPagesContainingMRI(handCuratedOverseasDomainsFile);
|
---|
147 |
|
---|
148 | // 0. get a list of all the web pages in the given domain listing where isMRI = true
|
---|
149 | ArrayList<Tuple> urlsList = getURLsForAllWebPagesInSiteListing(MongoDBQueryer.IS_MRI, domainsFile);
|
---|
150 | // produceURLsForPagesInMRI(domainsFile);
|
---|
151 |
|
---|
152 | // 1. calculate the population size, N, the number of all webpages in the given domain
|
---|
153 | // site listing where isMRI = true.
|
---|
154 | int N_totalNumPages = urlsList.size();
|
---|
155 |
|
---|
156 | // 2. write all the URLs in urlsList to a file
|
---|
157 | //File outFolder = domainsFile.getParentFile();
|
---|
158 | String filterName = (filterType == MongoDBQueryer.IS_MRI) ? "isMRI" : "containsMRI";
|
---|
159 | File outFile = new File(outFolder, filterName+"_full_"+domainsFile.getName());
|
---|
160 |
|
---|
161 | writeURLsToFile(urlsList, outFile, N_totalNumPages);
|
---|
162 | System.out.println("Wrote out full listing of web page URLs for sites in input domainsFile"
|
---|
163 | + "\ninto file: " + Utility.getFilePath(outFile));
|
---|
164 |
|
---|
165 | // 3. calculate sample size n for population size N if using 90% confidence and 5% margin of error
|
---|
166 | int n_numSampleURLs = calcSampleSize(N_totalNumPages);
|
---|
167 |
|
---|
168 | System.err.println("*** N, total number of web pages for which " + filterName + "=true from domain shortlist file: " + N_totalNumPages);
|
---|
169 | System.err.println(" (out of " + mongodbQueryer.countOfWebpagesMatching(filterType)
|
---|
170 | + " web pages across ALL sites for which " + filterName + " = true)");
|
---|
171 | System.err.println("*** n, sample size of web page URLs: " + n_numSampleURLs);
|
---|
172 |
|
---|
173 | // 4. Shuffle all the URLs and write the first n (sample size) URLs to a file
|
---|
174 | // Using a constant seed for reproducibility
|
---|
175 | // https://stackoverflow.com/questions/6284589/setting-a-seed-to-shuffle-arraylist-in-java-deterministically
|
---|
176 | Collections.shuffle(urlsList, new Random(FIXED_SEED));
|
---|
177 |
|
---|
178 | outFile = new File(outFolder, "random"+n_numSampleURLs+"_"+domainsFile.getName());
|
---|
179 | writeURLsToFile(urlsList, outFile, n_numSampleURLs);
|
---|
180 | System.out.println("Wrote a sample of n=" + n_numSampleURLs + " of web page URLs "
|
---|
181 | + "for the sites in input domainsFile\ninto file: " + Utility.getFilePath(outFile));
|
---|
182 |
|
---|
183 | // For N = 6557, z-alpha-over-2 = 1.6449 and m = 0.05 (5%),
|
---|
184 | // n = (z-alpha-over-2^2 x N) / (z-alpha-over-2^2 + 4 x (N-1) x m^2)
|
---|
185 | // = (1.6449^2Ã6557) ÷ (1.6449^2 + 4 à 6556Ã0.05^2) = 259.88526851 => 260 rounded up. Check.
|
---|
186 | }
|
---|
187 |
|
---|
188 | /**
|
---|
189 | * Calculates sample size n for binary outcomes at 90% confidence and 5% margine of error
|
---|
190 | * for given population size N.
|
---|
191 | * @return n, the sample size.
|
---|
192 | */
|
---|
193 | public int calcSampleSize(int N) {
|
---|
194 |
|
---|
195 | // calculate sample size n for population size N if using 90% confidence and 5% margin of error
|
---|
196 | // https://stats.stackexchange.com/questions/207584/sample-size-choice-with-binary-outcome
|
---|
197 | // https://www.statisticshowto.datasciencecentral.com/probability-and-statistics/find-sample-size/#CI1
|
---|
198 | // https://www.statisticshowto.datasciencecentral.com/z-alpha2-za2/
|
---|
199 |
|
---|
200 | double m = 0.05; // margin of error = 5%
|
---|
201 | // for 90% confidence, alpha is the remainder = 10% and alpha/2 = 5%.
|
---|
202 | // For 90% confidence, use the table of known z_alpha/2 values from step 1 of
|
---|
203 | // https://www.statisticshowto.datasciencecentral.com/z-alpha2-za2/
|
---|
204 | double z_alpha_over_2 = 1.6449;
|
---|
205 |
|
---|
206 | // Formula: n = (zalpha2 ^ 2 * N) / ((z-alpha-2 ^ 2) + 4(N-1)*m^2)
|
---|
207 | // see https://stats.stackexchange.com/questions/207584/sample-size-choice-with-binary-outcome
|
---|
208 | double n = (Math.pow(z_alpha_over_2, 2.0) * N) / (Math.pow(z_alpha_over_2, 2.0) + (4 * (N-1) * Math.pow(m,2.0)));
|
---|
209 |
|
---|
210 | // Round up to get a whole number:
|
---|
211 | return (int)Math.ceil(n);
|
---|
212 | }
|
---|
213 |
|
---|
214 | /**
|
---|
215 | * Writes out the first n URLs in urlsList into outFile.
|
---|
216 | */
|
---|
217 | private void writeURLsToFile(ArrayList<Tuple> urlsList, File outFile, final int n) {
|
---|
218 | try (
|
---|
219 | Writer writer = new BufferedWriter(new FileWriter(outFile));
|
---|
220 | ) {
|
---|
221 |
|
---|
222 | for (int i=0; i < n; i++) {
|
---|
223 | Tuple urlInfo = urlsList.get(i);
|
---|
224 |
|
---|
225 | //System.out.println(list.get(i));
|
---|
226 | writer.write(urlInfo + "\n"); // calls toString() on tuple of url -> countryCode
|
---|
227 | }
|
---|
228 | } catch(Exception e) {
|
---|
229 | logger.error("Unable to write to file " + Utility.getFilePath(outFile));
|
---|
230 | logger.error(e.getMessage(), e);
|
---|
231 | }
|
---|
232 | }
|
---|
233 |
|
---|
234 | /* ---------------------------------------- */
|
---|
235 | /**
|
---|
236 | * Create the file 5counts_containsMRISites_allNZGrouped.json
|
---|
237 | * that contains the count and domains for NZ sites (NZ origin or nz TLD) with pages
|
---|
238 | * that CONTAIN_MRI, followed by counts and domains listing for overseas sites
|
---|
239 | * that CONTAIN_MRI.
|
---|
240 | * @return full path of file generated
|
---|
241 | */
|
---|
242 | public String writeContainsMRISites_nzSitesAndTLDsGrouped() {
|
---|
243 |
|
---|
244 | File outFile = new File(outFolder, "5counts_containsMRISites_allNZGrouped.json");
|
---|
245 |
|
---|
246 | String filename = Utility.getFilePath(outFile);
|
---|
247 |
|
---|
248 | try (
|
---|
249 | Writer writer = new BufferedWriter(new FileWriter(outFile));
|
---|
250 | ) {
|
---|
251 | // first write out NZ sites and .nz TLD count and domains
|
---|
252 | mongodbQueryer.aggregateContainsMRIForNZ(writer, MongoDBQueryer.CONTAINS_MRI);
|
---|
253 | // next write out all overseas sites (not NZ origin or .nz TLD)
|
---|
254 | // that have no "mi" in the URL path as mi.* or */mi
|
---|
255 | boolean isMiInURLPath = false;
|
---|
256 | mongodbQueryer.aggregateContainsMRIForOverseas(writer, MongoDBQueryer.CONTAINS_MRI);
|
---|
257 |
|
---|
258 | } catch(Exception e) {
|
---|
259 | logger.error("Unable to write to file " + filename);
|
---|
260 | logger.error(e.getMessage(), e);
|
---|
261 | }
|
---|
262 |
|
---|
263 | System.err.println("*** Wrote file: " + filename);
|
---|
264 |
|
---|
265 | return filename;
|
---|
266 | }
|
---|
267 |
|
---|
268 | /**
|
---|
269 | * Create the file 5a_counts_tentativeNonAutotranslatedSites.json
|
---|
270 | * that contains the count and domains for NZ sites (NZ origin or nz TLD) that CONTAIN_MRI
|
---|
271 | * followed by counts and domain listing for overseas sites that are either from Australia
|
---|
272 | * or don't contain mi in their URL path.
|
---|
273 | * @return full path of file generated
|
---|
274 | */
|
---|
275 | public String writeTentativeNonAutotranslatedSites() {
|
---|
276 |
|
---|
277 | File outFile = new File(outFolder, "5a_counts_tentativeNonAutotranslatedSites.json");
|
---|
278 |
|
---|
279 | String filename = Utility.getFilePath(outFile);
|
---|
280 |
|
---|
281 | try (
|
---|
282 | Writer writer = new BufferedWriter(new FileWriter(outFile));
|
---|
283 | ) {
|
---|
284 | // first write out NZ sites and .nz TLD count and domains
|
---|
285 | mongodbQueryer.aggregateContainsMRIForNZ(writer, MongoDBQueryer.CONTAINS_MRI);
|
---|
286 | // next write out all overseas sites (not NZ origin or .nz TLD)
|
---|
287 | // that have no "mi" in the URL path as mi.* or */mi
|
---|
288 | boolean isMiInURLPath = false;
|
---|
289 | mongodbQueryer.aggregateContainsMRIForOverseas(writer, MongoDBQueryer.CONTAINS_MRI, isMiInURLPath);
|
---|
290 |
|
---|
291 | } catch(Exception e) {
|
---|
292 | logger.error("Unable to write to file " + filename);
|
---|
293 | logger.error(e.getMessage(), e);
|
---|
294 | }
|
---|
295 |
|
---|
296 | System.err.println("*** Wrote file: " + filename);
|
---|
297 |
|
---|
298 | return filename;
|
---|
299 | }
|
---|
300 |
|
---|
301 | /**
|
---|
302 | * Create the file 5b_counts_overseasSitesWithMiInPath.json
|
---|
303 | * Listing of the remainder of overseas sites that CONTAIN_MRI not included by
|
---|
304 | * writeTentativeNonAutotranslatedSites(): those that have mi in their URL path.
|
---|
305 | * This listing is separate to allow easier weeding out of product sites/autotranslated
|
---|
306 | * sites when eyeballing the listing output.
|
---|
307 | */
|
---|
308 | public String writeOverseasSitesWithMiInURLPath() {
|
---|
309 | File outFile = new File(outFolder, "5b_counts_overseasSitesWithMiInPath.json");
|
---|
310 |
|
---|
311 | String filename = Utility.getFilePath(outFile);
|
---|
312 | try (
|
---|
313 | Writer writer = new BufferedWriter(new FileWriter(outFile));
|
---|
314 | ) {
|
---|
315 | boolean isMiInURLPath = true;
|
---|
316 | mongodbQueryer.aggregateContainsMRIForOverseas(writer, MongoDBQueryer.CONTAINS_MRI, isMiInURLPath);
|
---|
317 |
|
---|
318 | } catch(Exception e) {
|
---|
319 | logger.error("Unable to write to file " + filename);
|
---|
320 | logger.error(e.getMessage(), e);
|
---|
321 | }
|
---|
322 |
|
---|
323 | System.err.println("*** Wrote file: " + filename);
|
---|
324 | return filename;
|
---|
325 | }
|
---|
326 |
|
---|
327 | public static void printUsage() {
|
---|
328 | System.err.println("Usage: SummaryTool [domains.txt]");
|
---|
329 | }
|
---|
330 |
|
---|
331 | /**
|
---|
332 | * If no args are passed in, generates complete containsMRI file listings for NZ and overseas web SITES (domains),
|
---|
333 | * with overseas web sites that have mi (mi.* or *\/mi) in the URL path listed separately.
|
---|
334 | * You can then manually inspect the domains in this listing to shortlist which of these sites are not automatically
|
---|
335 | * translated and really contain at least one webpage containing at least one sentence in MRI.
|
---|
336 | * If a file is passed in containing a list of domains, then this first generates a full listing of all webpages
|
---|
337 | * matching isMRI for each site in the domain list. It then generates a smaller set of random webpages matching
|
---|
338 | * isMRI for the pooled sites in the domain list where the sample size of URLs produced is sufficient for giving
|
---|
339 | * 90% confidence with 5% margin of error for testing binary outcomes, see
|
---|
340 | * https://stats.stackexchange.com/questions/207584/sample-size-choice-with-binary-outcome
|
---|
341 | */
|
---|
342 | public static void main(String args[]) {
|
---|
343 | if(args.length >= 2) {
|
---|
344 | printUsage();
|
---|
345 | System.exit(-1);
|
---|
346 | }
|
---|
347 |
|
---|
348 | try (
|
---|
349 | MongoDBQueryer mongodb = new MongoDBQueryer();
|
---|
350 | ) {
|
---|
351 |
|
---|
352 | mongodb.connectToDB();
|
---|
353 |
|
---|
354 | // output files will be stored in mongodb-data-auto
|
---|
355 | File outFolder = new File("../mongodb-data-auto/").getAbsoluteFile();
|
---|
356 | SummaryTool listing = new SummaryTool(mongodb, outFolder);
|
---|
357 |
|
---|
358 | System.out.println("*************************************");
|
---|
359 |
|
---|
360 |
|
---|
361 | if(args.length >= 1) { // web page urls listing mode
|
---|
362 | File domainsFile = new File(args[0]);
|
---|
363 | if(!domainsFile.exists()) {
|
---|
364 | System.err.println("File " + domainsFile + " does not exist");
|
---|
365 | System.exit(-1);
|
---|
366 | }
|
---|
367 |
|
---|
368 | //String isMRIFile = listing.produceURLsForPagesInMRI(domainsFile);
|
---|
369 | //String containsMRIFile = listing.produceURLsForPagesContainingMRI(domainsFile);
|
---|
370 |
|
---|
371 |
|
---|
372 | // TODO: for NZ, do IS_MRI. For overseas still CONTAINS_MRI
|
---|
373 | // then also do the shuffle to gen X num of random web page URLs.
|
---|
374 | //String filename = listing.webPagesOfAllNZSitesAndDomainListing(domainsFile);
|
---|
375 | listing.mriWebPageListingForDomainListing(domainsFile);
|
---|
376 |
|
---|
377 | // TODO: generate the special table (6)
|
---|
378 |
|
---|
379 | } else {
|
---|
380 |
|
---|
381 | // calculating sample size works:
|
---|
382 | //System.err.println("For N = " + 4360 + ", n = " + listing.calcSampleSize(4360));
|
---|
383 | //System.err.println("For N = " + 681 + ", n = " + listing.calcSampleSize(681));
|
---|
384 |
|
---|
385 | // get all sites where >0 pages have containsMRI=true
|
---|
386 | // grouping NZ sites and .nz TLDs together and remainder under overseas
|
---|
387 | // geolocations.
|
---|
388 | String filename = listing.writeContainsMRISites_nzSitesAndTLDsGrouped();
|
---|
389 |
|
---|
390 | // separately:
|
---|
391 | // - all NZ containsMRI + overseas tentative non-product sites with containMRI
|
---|
392 | // - overseas tentative product sites with containMRI
|
---|
393 | filename = listing.writeTentativeNonAutotranslatedSites();
|
---|
394 | filename = listing.writeOverseasSitesWithMiInURLPath();
|
---|
395 |
|
---|
396 | // TODO: generate the tables
|
---|
397 |
|
---|
398 | String[] tableFileNames = mongodb.writeTables(outFolder);
|
---|
399 | // for each table file name, generate the geojson-features .json file
|
---|
400 | // that GEOJSON_MAP_TOOL_URL takes as input to produce a map.
|
---|
401 |
|
---|
402 | for(int i = 1; i < tableFileNames.length; i++) { // empty element at 0
|
---|
403 | String tablefilename = tableFileNames[i] + ".json"; // filenames have no suffix
|
---|
404 |
|
---|
405 | File countsTableFile = new File(outFolder, tablefilename);
|
---|
406 | if(!countsTableFile.exists()) {
|
---|
407 | logger.error("@@@ File " + countsTableFile + " does not exist!");
|
---|
408 | logger.error("@@@ Can't generate map date for this.");
|
---|
409 | continue;
|
---|
410 | }
|
---|
411 | String countsTableFilename = outFolder + File.separator + tablefilename;
|
---|
412 | CountryCodeCountsMapData mapData
|
---|
413 | = new CountryCodeCountsMapData(countsTableFilename);
|
---|
414 | String geoJsonFilename = mapData.writeFeaturesGeoJsonToFile(CountryCodeCountsMapData.SUPPRESS_MAPDATA_DISPLAY);
|
---|
415 |
|
---|
416 |
|
---|
417 | /*
|
---|
418 | // Ensure the geo-json file generated exists
|
---|
419 | //String geoJsonFilename = outFolder + File.separator
|
---|
420 | //+ GEOJSON_FEATURES_FILE_PREFIX + tablefilename;
|
---|
421 |
|
---|
422 | File geoJsonFile = new File(geoJsonFilename);
|
---|
423 | if(!geoJsonFile.exists()) {
|
---|
424 | System.err.println("@@@ geoJson file " + geoJsonFilename + " not generated!");
|
---|
425 | continue;
|
---|
426 | }
|
---|
427 | */
|
---|
428 | System.err.println("**** Wrote mapdata to file " + geoJsonFilename);
|
---|
429 | //System.err.println(" Paste the file's geojson contents into "
|
---|
430 | //+ "the editor at " + CountryCodeCountsMapData.GEOJSON_MAP_TOOL_URL
|
---|
431 | //+ " to see the data arranged on a world map");
|
---|
432 | System.err.println("Total count for query: " + mapData.getTotalCount());
|
---|
433 |
|
---|
434 | // Running the command:
|
---|
435 | // python -mjson.tool file.json
|
---|
436 | // to confirm if the JSON in the provided json file parses/is valid JSON.
|
---|
437 |
|
---|
438 | String[] cmdArgs = {"python", "-mjson.tool", geoJsonFilename};
|
---|
439 | SafeProcess p = new SafeProcess(cmdArgs);
|
---|
440 | int retVal = p.runProcess();
|
---|
441 | logger.info("\nJson validation tool returned with value: " + retVal);
|
---|
442 | if(retVal == 0) {
|
---|
443 | logger.info("Valid geo json.");
|
---|
444 | } else {
|
---|
445 | logger.error(" std out: " + p.getStdOutput());
|
---|
446 | logger.error(" std err: " + p.getStdError());
|
---|
447 | logger.error("INVALID geo json generated. Not attempting screenshot of map.");
|
---|
448 | continue;
|
---|
449 | }
|
---|
450 |
|
---|
451 |
|
---|
452 | /*boolean uriEncoded = true;
|
---|
453 | String mapDataEncodedStr = mapData.getFeaturesGeoJsonString(uriEncoded);
|
---|
454 | System.err.println("Encoded string: " + mapDataEncodedStr);
|
---|
455 | */
|
---|
456 |
|
---|
457 | System.err.println("Data URL string: " + mapData.getAsMapURL());
|
---|
458 | System.err.println();
|
---|
459 | mapData.geoJsonMapScreenshot(outFolder, tablefilename);
|
---|
460 | System.err.println("---");
|
---|
461 |
|
---|
462 | // TODO: breaks after first table -> map conversion
|
---|
463 | break;
|
---|
464 | }
|
---|
465 |
|
---|
466 | }
|
---|
467 |
|
---|
468 | System.out.println("*************************************");
|
---|
469 | } catch(Exception e) {
|
---|
470 | logger.error(e.getMessage(), e);
|
---|
471 | }
|
---|
472 | }
|
---|
473 | }
|
---|