Changeset 1399 in lmdz_wrf for trunk/tools

Timestamp:

Jan 9, 2017, 2:39:54 PM (8 years ago)

Author:

lfita

Message:

Adding `exctcurvloc', exact location on the i,j grid space of the curve in 'curvelocalize_2D'
Fixing right weights in 'linearint_3x3weights'
num_ordinal: Function to provide the ordinal of a given number, language, format and gender
removing error message on `reduce_last_spaces' when no space is found
adding some new signs in `ASCII_to'

File:

• trunk/tools/generic_tools.py (modified) (10 diffs)

trunk/tools/generic_tools.py

@@ -89 +89 @@
 # multi_index_string: Function to provide the indeces of every repetition of a group of characters within a string
 # num_chainSnum: Function to pass a value to a `ChainStrNum' number
+# num_ordinal: Function to provide the ordinal of a given number, language, format and gender
 # num_split: Function to split a string at each numeric value keeping the number as the last character of each cut
 # oper_submatrix: Function to perform an operation of a given matrix along a sub-set of values along its dimensions
@@ -6714 +6715 @@
 
     Lstring = len(string)
+    if Lstring < 1:
+        print errormsg
+        print '  ' + fname + ": String '" + string + "' too short !!"
+        quit(-1)
 
     firstspace = -1
@@ -6723 +6728 @@
 
     if firstspace == -1:
-        print errormsg
-        print '  ' + fname + ": string '" + string + "' is not ended by spaces!!"
-        print '    char. number of last character:',ord(string[Lstring:Lstring+1])
-        quit(-1)
+        firstspace = Lstring + 1
 
     newstr = string[0:firstspace]
@@ -9541 +9543 @@
     langs = ['tex']
 
+    #From: http://tex.stackexchange.com/questions/14872/croatian-serbian-letters-in-cv-problem
+    # From: https://www.dpg-tagung.de/dd14/infochars.html?language=en
     texchars = {'\\': '\\textbackslash', \
       'á': "\\'a", 'é': "\\'e", 'í': "\\'i", 'ó': "\\'o", 'ú': "\\'u",               \
+      'ć': "\\'c", 'ǵ': "\\'g", 'áž±': "\\'k", 'ĺ': "\\'l", 'áž¿': "\\'m", 'ń': "\\'n",  \
+      'ṕ': "\\'p", 'ŕ': "\\'r", 'ś': "\\'s", 'ẃ': "\\'w", 'Ãœ': "\\'y", 'ź': "\\'z",  \
       'à ': "\\`a", 'Ú': "\\`e", 'ì': "\\`i", 'ò': "\\`o", 'ù': "\\`u",               \
+      'ǹ': "\\`n", 'ẁ': "\\`w", 'ỳ': "\\`y",                                         \
       'â': "\\^a", 'ê': "\\^e", 'î': "\\^i", 'Î': "\\^o", 'û': "\\^u",               \
+      'ĉ': "\\^c", 'ĝ': "\\^g", 'ĥ': "\\^h", 'ĵ': "\\^j", 'ŝ': "\\^s", 'ŵ': "\\^w",  \
+      'ŷ': "\\^y", 'ẑ': "\\^z",                                                      \
       'À': '\\"a', 'ë': '\\"e', 'ï': '\\"i', 'ö': '\\"o', 'Ì': '\\"u',               \
-      'ç': '\c{c}', 'ñ': '\~{n}',                                                    \
+      'ឧ': '\\"h', 'ẗ': '\\"t', 'áº
+': '\\"w', 'ẍ': '\\"x', 'ÿ': '\\"y',               \
+      'ă': '\\u{a}', 'ĕ': '\\u{e}', 'ĭ': '\\u{i}', 'ŏ': '\\u{o}', 'ŭ': '\\u{u}',     \
+      'ğ': '\\u{g}',                                                                 \
+      'ǎ': '\\v{a}', 'ě': '\\v{e}', 'ǐ': '\\v{i}', 'ǒ': '\\v{o}', 'ǔ': '\\v{u}',     \
+      'č': '\\v{c}', 'ď': '\\v{d}', 'ǧ': '\\v{g}', 'ȟ': '\\v{h}', 'ǰ': '\\v{j}',     \
+      'Ç©': '\\v{k}', 'ÄŸ': '\\v{l}', 'ň': '\\v{n}', 'ř': '\\v{r}', 'Å¡': '\\v{s}',     \
+      'Å¥': '\\v{t}', 'ÅŸ': '\\v{z}',                                                  \
+      'å': '\\a{}', 'ů': '\\r{u}', 'ẘ': '\\r{w}', 'ẙ': '\\r{y}',                     \
+      'ã': '\\~{a}', 'ẜ': '\\~{e}', 'ĩ': '\\~{i}', 'õ': '\\~{o}', 'ũ': '\\~{u}',     \
+      'ñ': '\\~{n}', 'Ṝ': '\\~{v}', 'ỹ': '\\~{y}',                                   \
+      'ç': '\\c{c}', 'ᾑ': '\\c{d}', 'È©': '\\c{e}', 'Ä£': '\\c{g}', 'áž©': '\\c{h}',     \
+      'Ä·': '\\c{k}', 'ÄŒ': '\\c{l}', 'ņ': '\\c{n}', 'ŗ': '\\c{r}', 'ş': '\\c{s}',     \
+      'Å£': '\\c{t}',                                                                 \
+      'ł': '\\l{}', 'Þ': '\\o{}', 'đ': '\\dj', 'ß': '\\ss{}',                                     \
       'Á': "\\'A", 'É': "\\'E", 'Í': "\\'I", 'Ó': "\\'O", 'Ú': "\\'U",               \
+      'Ć': "\\'C", 'ÇŽ': "\\'G", 'áž°': "\\'K", 'Ĺ': "\\'L", 'ស': "\\'M", 'Ń': "\\'N",  \
+      'Ṕ': "\\'P", 'Ŕ': "\\'R", 'Ś': "\\'S", 'Ẃ': "\\'W", 'Ý': "\\'Y", 'Ź': "\\'Z",  \
       'À': "\\`A", 'È': "\\`E", 'Ì': "\\`I", 'Ò': "\\`O", 'Ù': "\\`U",               \
+      'Çž': "\\`N", 'Ẁ': "\\`W", 'Ỳ': "\\`Y",                                         \
       'Â': "\\^A", 'Ê': "\\^E", 'Î': "\\^I", 'Ô': "\\^O", 'Û': "\\^U",               \
+      'Ĉ': "\\^C", 'Ĝ': "\\^G", 'Ä€': "\\^H", 'ÄŽ': "\\^J", 'Ŝ': "\\^S", 'ÅŽ': "\\^W",  \
+      'Ŷ': "\\^Y", 'Ẑ': "\\^Z",                                                      \
       'Ä': '\\"A', 'Ë': '\\"E', 'Ï': '\\"I', 'Ö': '\\"O', 'Ü': '\\"U',               \
-      'Ç': '\\c{C}', 'Ñ': '\\~{N}',                                                  \
+      'ដ': '\\"H', 'Ẅ': '\\"W', 'Ẍ': '\\"X', 'Åž': '\\"Y',                            \
+      'Ă': '\\u{A}', 'Ĕ': '\\u{E}', 'Ĭ': '\\u{I}', 'Ŏ': '\\u{O}', 'Ŭ': '\\u{U}',     \
+      'Ğ': '\\u{G}', 'Ŝ': '\\u{Z}',                                                  \
+      'Ǎ': '\\v{A}', 'Ě': '\\v{E}', 'Ǐ': '\\v{I}', 'Ǒ': '\\v{O}', 'Ǔ': '\\v{U}',     \
+      'Č': '\\v{C}', 'Ď': '\\v{D}', 'ÇŠ': '\\v{G}', 'Ȟ': '\\v{H}', 'Çš': '\\v{K}',     \
+      'Äœ': '\\v{L}', 'Ň': '\\v{N}', 'Ř': '\\v{R}', 'Å ': '\\v{S}', 'Å€': '\\v{T}',     \
+      'Ŝ': '\\v{Z}',                                                                 \
+      'Ã
+': '\\AA{}', 'Å®': '\\r{U}',                                                  \
+      'Ã': '\\~{A}', 'Ẍ': '\\~{E}', 'Äš': '\\~{I}', 'Õ': '\\~{O}', 'Åš': '\\~{U}',     \
+      'Ñ': '\\~{N}', 'Ṍ': '\\~{V}', 'Ở': '\\~{Y}',                                   \
+      'Ç': '\\c{C}', 'ថ': '\\c{D}', 'Èš': '\\c{E}', 'Ä¢': '\\c{G}', 'ážš': '\\c{H}',     \
+      'Ķ': '\\c{K}', 'Ä»': '\\c{L}', 'Å
+': '\\c{N}', 'Ŗ': '\\c{R}', 'Ş': '\\c{S}',     \
+      'Å¢': '\\c{T}',                                                                 \
+      'Ł': '\\L{}', 'Ø': '\O{}', 'Đ': '\\DJ',                                        \
       '¡': '!`', '¿': '¿`', '%': '\\%', '#': '\\#', '&': '\\&', '$': '\\$',          \
-      '_': '\\_', '·': '\\textperiodcentered', '<': '$<$', '>': '$>$',               \
-      '': '*', 'ª': '$^{a}$', 'º': '$^{o}$', '°': '$^{\\circ}$', '\n': '\\\\\n',   \
-      '\t': '\\medskip', '“': '``', '”': '\'\'', '^': '\^'}
+      '_': '\\_', '–': '--', '·': '\\textperiodcentered', '<': '$<$', '>': '$>$',    \
+      '': '*', 'ª': '$^{a}$', 'º': '$^{o}$', '°': '$^{\\circ}$', '²': '$^{2}$',    \
+      '\n': '\\\\\n', '\t': '\\medskip',                                             \
+      '“': '``', '”': '\'\'', '^': '\^', '’': "'",                                   \
+      '≥': '$\geq$', '~': '$\sim$'}
 
     if to == 'tex':
@@ -10927 +10972 @@
     # Normalizing values (only the first 4)
     if iijj >= 3:
-        inc = np.sum(sortdist[0:4])
+        inc = np.sum(1./np.array(sortdist[0:4]))
         maxdist = sortdist[3]
         Nvals = 4
     else:
-        inc = np.sum(sortdist[0:iijj])
+        inc = np.sum(1./np.array(sortdist[0:iijj]))
         maxdist = sortdist[iijj]
         Nvals = iijj
 
-    wgt = np.where(dist <= maxdist, dist / inc, 0.)
+    wgt = np.where(dist <= maxdist, 1./dist/inc, 0.)
+    # Case with more than 4 points with less distance than `maxdist' (point at the diagonal)
     if np.sum(wgt) > 1.000001:
+        print 'maxdist:', maxdist
         wgt = np.where(dist <= maxdist, dist, 0.)
-        inc = np.sum(wgt)
-        wgt = np.where(wgt > 0., wgt/inc, 0.)
+        inc = 0.
+        wgtflat = wgt.flatten()
+        for i in range(9):
+            if wgtflat[i] > 0.: inc = inc + 1./wgtflat[i]
+        wgt = np.where(wgt > 0., 1./wgt/inc, 0.)
         Nvals = np.sum(wgt > 0.)
+
+    if np.all(wgt == 0.):
+        print errormsg
+        print '  ' + fname + ': no point found !!'
+        print '    point to interpolate:', intval0, ' around x & y & dist _____'
+        for i in range(3):
+            print xvals[:,i], '&', yvals[:,i], '&', dist[:,i]
+        print '    sortdist:', sortdist, 'weights _____'
+        for i in range(3):
+            print wgt[:,i]
+        quit(-1)
 
 #    print 'Lluis weights _______', sortdist[0:4],'max:',maxdist,'sum:',np.sum(wgt)
@@ -10954 +11015 @@
              x,y_j,i       x,y_j,i+1
 
-                curve(t)
-
-                           curve(t+1)
+                curve(k)
+
+                           curve(k+1)
 
              x,y_j-1,i     x,y_j-1,i+1
@@ -10972 +11033 @@
         Nwgt= Number of weights
         ijaorund3x3= [Npts,2,9] distance (in xval, yval units) to the curve position
+        exctcurvloc= [Npts,2] matrix with the exact i,j grid point in the 2D space of the field positions
     """
     fname = 'curvelocalize_2D'
@@ -10979 +11041 @@
     curvloc = np.ones((curve.shape), dtype=int)*(-1)
     Nwgts = np.zeros((Npts), dtype=int)
-    curveweights = np.zeros((Npts,3,3), dtype=float)
+    curveweights = np.zeros((Npts,3,3), dtype=np.float)
     ijaround3x3 = np.zeros((Npts,2,9), dtype=int)
+    exctcurvloc = np.ones((curve.shape), dtype=np.float)*(-1)
 
     for icv in range(Npts):
@@ -10999 +11062 @@
         yaroundvls = vals_around(yvals,ijloc)
 
-        curveweights[icv,:,:], Nwgts[icv], ijaround3x3[icv,:,:,] =                   \
+        curveweights[icv,:,:], Nwgts[icv], ijaround3x3[icv,:,:] =                    \
           linearint_3x3weights(xaroundvls, yaroundvls, curve[icv,:])
-        
-    return curvloc, curveweights, Nwgts, ijaround3x3
-
+        exctcurvloc[icv,:] = curvloc[icv,:]
+
+        for ip in range(Nwgts[icv]):
+            ipy = ijaround3x3[icv,0,ip]
+            ipx = ijaround3x3[icv,1,ip]
+            exctcurvloc[icv,0] = np.float(exctcurvloc[icv,0]) + curveweights[icv,ipy,ipx]*(ipy-1.)
+            exctcurvloc[icv,1] = np.float(exctcurvloc[icv,1]) + curveweights[icv,ipy,ipx]*(ipx-1.)
+            print '    ' + fname + 'ip:', ip, 'ipy:', ipy-1., 'wgt:', curveweights[icv,ipy,ipx], \
+              'exctcurv:', exctcurvloc[icv,0]
+        print '  ' + fname + ' icv:', icv, 'curve:', curve[icv,:], 'ijloc:', ijloc, \
+          'cwgt:', curveweights[icv,:,:], 'exctloc:', exctcurvloc[icv,:]
+#    quit()
+
+    return curvloc, curveweights, Nwgts, ijaround3x3, exctcurvloc
+#
 #lon1d = np.arange(10)
 #lat1d = np.arange(-5,5)
 #dx = lon1d.shape[0]
 #dy = lat1d.shape[0]
-
+#
 #lons = np.zeros((dy,dx), dtype=np.float)
 #lats = np.zeros((dy,dx), dtype=np.float)
-
+#
 #for iy in range(dy):
 #    lons[iy,:] = lon1d
 #for ix in range(dx):
 #    lats[:,ix] = lat1d
-
+#
 #Ncurve = 12
 #iicurve = 0
@@ -11028 +11103 @@
 #Dxcurve = (eloncurve-iloncurve)/(Ncurve-1.)
 #Dycurve = (elatcurve-ilatcurve)/(Ncurve-1.)
-
+#
 #curve = np.zeros((Ncurve,2), dtype=np.float)
 #for icv in range(Ncurve):
 #    curve[icv,0] = ilatcurve + Dycurve*icv
 #    curve[icv,1] = iloncurve + Dxcurve*icv
-
+#
 #print 'Curve origin:', ilatcurve, ',', iloncurve
 #print 'Curve ending:', elatcurve, ',', eloncurve
-
+#
 #print curve
-
-#loccurve, wgts, Nwgts, loc3x3 = curvelocalize_2D(curve,lons,lats)
+#
+#loccurve, wgts, Nwgts, loc3x3, exctloccurve = curvelocalize_2D(curve,lons,lats)
 #for ipt in range(Ncurve):
-#    print 'pos:', ipt, 'curve:', curve[ipt,:],'loccurve:', loccurve[ipt,:], 'wgts:', wgts[ipt,:,:], 'Nwgts:', Nwgts, 'loc3x3:', loc3x3[ipt,:]
+#    print 'pos:', ipt, 'curve:', curve[ipt,:],'loccurve:', loccurve[ipt,:], 'wgts:', wgts[ipt,:,:], 'Nwgts:', Nwgts[ipt], 'loc3x3:', loc3x3[ipt,:], 'exctloccurve:', exctloccurve[ipt,:]
+#    for iw in range(Nwgts[ipt]):
+#        iwy = loc3x3[ipt,0,iw]
+#        iwx = loc3x3[ipt,1,iw]
+#        print '   sort wgts:', iw,' iwy,iwx:', iwy,',',iwx, 'wgt:', wgts[ipt,iwy,iwx]
+
+def num_ordinal(num, lang='eng', kind='numOr', fmt='ascii', gender='masc'):
+    """ Function to provide the ordinal of a given number, language, format and gender
+      num= number to provide the ordinal
+      lang= language of the ordinal
+        'eng': English
+      kind= kind of the ordinal
+        'numOr': as num[Ord]
+        'Ordinal': long ordinal (not fully ready)
+      fmt= format of the ordinal
+        'ascii': standard ASCII
+        'utf': utf-8 (with accents)
+      gender= gender of the ordinal
+        'fem': femenin
+        'masc': masculin
+      >>> num_ordinal(2, 'eng', 'numOr', 'ascii')
+      '2nd'
+      >>> num_ordinal(2, 'eng', 'Ordinal', 'ascii')
+      'second'
+      >>> num_ordinal(7, 'cat', 'Ordinal', 'utf')
+      'setÚ'
+      >>> num_ordinal(1, 'cat', 'Ordinal', 'utf', fem)
+      'primera'
+    """
+    fname = 'num_ordinal'
+
+    languages = ['cat', 'eng']
+    kinds = ['numOr', 'Ordinal']
+    formats = ['ascii', 'utf']
+    genders = ['fem', 'masc']
+
+    if not searchInlist(languages, lang):
+        print errormsg
+        print '  ' + fname + ": language '" + lang + "' not ready !!"
+        print '    available ones:', languages
+        quit(-1)
+
+    if not searchInlist(kinds, kind):
+        print errormsg
+        print '  ' + fname + ": kind '" + kind + "' not ready !!"
+        print '    available ones:', kinds
+        quit(-1)
+
+    if not searchInlist(formats, fmt):
+        print errormsg
+        print '  ' + fname + ": format '" + fmt + "' not ready !!"
+        print '    available ones:', fomrats
+        quit(-1)
+
+    if not searchInlist(genders, gender):
+        print errormsg
+        print '  ' + fname + ": gender '" + gender + "' not ready !!"
+        print '    available ones:', genders
+        quit(-1)
+
+    # Catalan
+    # short ordinals  ([fem, masc])
+    catord = {1: ['era', 'er'], 2: ['ona', 'on'], 3: ['era', 'er'], 4: ['rta', 'rt']}
+    # Last special ordinal
+    catlastord = 4
+    # generic ordinal ([fem, masc])
+    catgenord = ['ena', 'Ú']
+    # long ordinals
+    catLord = {1: ['primera', 'primer'], 2: ['segona', 'segon'], 3: ['tercera',      \
+      'tercer'], 4: ['quarta', 'quart'], 5: ['cinquena', 'cinquÚ'], 6: ['sisena',    \
+      'sisÚ'], 7: ['setena', 'setÚ'], 8: ['vuitena', 'vuitÚ'], 9: ['novena', 'novÚ']}
+
+    # English 
+    # short ordinals ([fem, masc])
+    engord = {1: ['st', 'st'], 2: ['nd', 'nd'], 3: ['rd', 'rd']}
+    # Last special ordinal
+    englastord = 3
+    # generic ordinal ([fem, masc])
+    enggenord = ['th','th']
+    # long ordinals
+    engLord = {1: ['first', 'first'], 2: ['second', 'second'], 3: ['third', 'third'],\
+      4: ['fourth', 'fourth'], 5: ['fifth', 'fifth'],  6: ['sixth','sixth'],         \
+      7: ['seventh', 'seventh'], 8: ['eight', 'eight'], 9: ['nineth', 'nineth']}
+
+    if lang == 'cat':
+        ords = catord
+        lastord = catlastord
+        genord = catgenord
+        Lords = catLord
+    elif lang == 'eng':
+        ords = engord
+        lastord = englastord
+        genord = enggenord
+        Lords = engLord
+
+    Snum = str(num)
+    if num > 10:
+        LSnum = len(Snum)
+        Senu = Snum[LSnum:LSnum+1]
+    else:
+        Senu = Snum
+
+    # Ordinal
+    if num < lastord:
+       Ord = ords[num]
+    else:
+       Ord = genord 
+    if gender == 'fem': Sord = Ord[0]
+    else: Sord = Ord[1]
+
+    # Long ordinal
+    if num < 10:
+       LOrd = Lords[num]
+    else:
+       print warnmsg
+       print '  ' + fname + ' long ordinal not ready for:', num, '!!'
+       LOrd = 'None'
+    if gender == 'fem': SLord = LOrd[0]
+    else: SLord = LOrd[1]
+
+    if kind == 'numOr':
+        stringv = Sord
+    elif kind == 'Ordinal':
+        stringv = SLord
+    else:
+        print errormsg
+        print '  ' + fname + ": kind '" + kind + "' not ready !!"
+        print '    available ones:', kinds
+        quit(-1)
+
+    return stringv
 
 #quit()