replace-hue example from Ivan

diff --git a/share/examples/Makefile.am b/share/examples/Makefile.am
index 8a440c413007c2f9087870365bcbccc88eabdafd..a89246048c71e789f74dd556081d1b8010773017 100644 (file)
--- a/share/examples/Makefile.am
+++ b/share/examples/Makefile.am
@@ -31,6 +31,7 @@ example_DATA = \
        live-path-effects-pathalongpath.svg \
        filters.svg \
        svgfont.svg \
-       tref.svg
+       tref.svg \
+       replace-hue.svg
 
 EXTRA_DIST = $(example_DATA)

diff --git a/share/examples/replace-hue.svg b/share/examples/replace-hue.svg
new file mode 100644 (file)
index 0000000..791d124
--- /dev/null
+++ b/share/examples/replace-hue.svg
@@ -0,0 +1,504 @@
+<?xml version="1.0" encoding="UTF-8" standalone="no"?>
+<!-- Created with Inkscape (http://www.inkscape.org/) -->
+
+<svg xmlns:dc="http://purl.org/dc/elements/1.1/" xmlns:cc="http://creativecommons.org/ns#" xmlns:rdf="http://www.w3.org/1999/02/22-rdf-syntax-ns#" xmlns:svg="http://www.w3.org/2000/svg" xmlns="http://www.w3.org/2000/svg" xmlns:xlink="http://www.w3.org/1999/xlink" xmlns:sodipodi="http://sodipodi.sourceforge.net/DTD/sodipodi-0.dtd" xmlns:inkscape="http://www.inkscape.org/namespaces/inkscape" width="900" height="300" version="1.1" id="svg2" sodipodi:version="0.32" inkscape:version="0.47pre1">
+<sodipodi:namedview pagecolor="#ffffff" bordercolor="#666666" borderopacity="1" objecttolerance="10" gridtolerance="10" guidetolerance="10" inkscape:pageopacity="0" inkscape:pageshadow="2" inkscape:window-width="1920" inkscape:window-height="1170" id="namedview107" showgrid="false" inkscape:zoom="1" inkscape:cx="514.76695" inkscape:cy="-14.43644" inkscape:window-x="-5" inkscape:window-y="-3" inkscape:current-layer="layer1" inkscape:window-maximized="1" />
+<metadata id="metadata109">
+<rdf:RDF>
+<cc:Work rdf:about="">
+<dc:format>image/svg+xml</dc:format>
+<dc:type rdf:resource="http://purl.org/dc/dcmitype/StillImage" />
+<dc:title></dc:title>
+</cc:Work>
+</rdf:RDF>
+</metadata>
+<defs id="defs4">
+<g id="hueImage">
+<linearGradient id="gradient">
+<stop id="stop8" stop-color="#f00" offset="0.000" />
+<stop id="stop10" stop-color="#ff0" offset="0.167" />
+<stop id="stop12" stop-color="#0f0" offset="0.333" />
+<stop id="stop14" stop-color="#0ff" offset="0.500" />
+<stop id="stop16" stop-color="#00f" offset="0.667" />
+<stop id="stop18" stop-color="#f0f" offset="0.833" />
+<stop id="stop20" stop-color="#f00" offset="1.000" />
+</linearGradient>
+<rect style="fill:url(#gradient)" y="0" x="0" id="rect22" height="299" width="300" />
+</g>
+<filter color-interpolation-filters="sRGB" width="1" height="1" y="0" x="0" id="replaceHueFromLayer">
+<!-- This is mainly to make sure the flood color has the desired effect. It would be better to use "real" colors and compute the hue from them instead, and if so linearRGB might actually have an advantage. -->
+<!-- Set up p, q and q-p -->
+<feColorMatrix id="feColorMatrix25" values="1 0 0 0 0 1 0 0 0 0 1 0 0 0 0 0 0 0 0 1" type="matrix" result="r" in="SourceGraphic" />
+<feColorMatrix id="feColorMatrix27" values="0 1 0 0 0 0 1 0 0 0 0 1 0 0 0 0 0 0 0 1" type="matrix" result="g" in="SourceGraphic" />
+<feColorMatrix id="feColorMatrix29" values="0 0 1 0 0 0 0 1 0 0 0 0 1 0 0 0 0 0 0 1" type="matrix" result="b" in="SourceGraphic" />
+<feBlend id="feBlend31" result="minrg" in2="g" in="r" mode="darken" />
+<feBlend id="feBlend33" result="p" in2="b" in="minrg" mode="darken" />
+<feBlend id="feBlend35" result="maxrg" in2="g" in="r" mode="lighten" />
+<feBlend id="feBlend37" result="q" in2="b" in="maxrg" mode="lighten" />
+<feComposite k1="0" id="feComposite39" k4="1" k3="1" k2="-1" operator="arithmetic" result="pminq" in2="p" in="q" />
+<!-- p-q+1 = 1-(q-p), with the right alpha :) -->
+<feColorMatrix id="feColorMatrix41" values="-1 0 0 0 1 0 -1 0 0 1 0 0 -1 0 1 0 0 0 0 1" type="matrix" result="qminp" in="pminq" />
+<!-- Get hq-hp and hrgb-hp -->
+<feImage id="feImage43" result="hueImage" xlink:href="#hueImage" />
+<feColorMatrix id="feColorMatrix45" values="1 0 0 0 0 0 1 0 0 0 0 0 1 0 0 0 0 0 0 1" type="matrix" result="hrgb" in="hueImage" />
+<feColorMatrix id="feColorMatrix47" values="1 0 0 0 0 1 0 0 0 0 1 0 0 0 0 0 0 0 0 1" type="matrix" result="hr" in="hueImage" />
+<feColorMatrix id="feColorMatrix49" values="0 1 0 0 0 0 1 0 0 0 0 1 0 0 0 0 0 0 0 1" type="matrix" result="hg" in="hueImage" />
+<feColorMatrix id="feColorMatrix51" values="0 0 1 0 0 0 0 1 0 0 0 0 1 0 0 0 0 0 0 1" type="matrix" result="hb" in="hueImage" />
+<feBlend id="feBlend53" result="hminrg" in2="hg" in="hr" mode="darken" />
+<feBlend id="feBlend55" result="hp" in2="hb" in="hminrg" mode="darken" />
+<feBlend id="feBlend57" result="hmaxrg" in2="hg" in="hr" mode="lighten" />
+<feBlend id="feBlend59" result="hq" in2="hb" in="hmaxrg" mode="lighten" />
+<feComposite k1="0" id="feComposite61" k4="1" k3="1" k2="-1" operator="arithmetic" result="hpminhq" in2="hp" in="hq" />
+<!-- hp-hq+1 = 1-(hq-hp), with the right alpha :) -->
+<feColorMatrix id="feColorMatrix63" values="-1 0 0 0 1 0 -1 0 0 1 0 0 -1 0 1 0 0 0 0 1" type="matrix" result="hqminhp" in="hpminhq" />
+<feComposite k1="0" id="feComposite65" k4="1" k3="1" k2="-1" operator="arithmetic" result="hpminhrgb" in2="hp" in="hrgb" />
+<!-- hp-hrgb+1 = 1-(hrgb-hp), with the right alpha :) -->
+<feColorMatrix id="feColorMatrix67" values="-1 0 0 0 1 0 -1 0 0 1 0 0 -1 0 1 0 0 0 0 1" type="matrix" result="hrgbminhp" in="hpminhrgb" />
+<!-- Compute (hrgb-hp)/(hq-hp) -->
+<feComponentTransfer id="feComponentTransfer69" result="invhqminhp" in="hqminhp">
+<!-- Computes (1/10)*(1/(hq-hp)) -->
+<feFuncR id="feFuncR71" exponent="-1" amplitude="0.1" type="gamma" />
+<feFuncG id="feFuncG73" exponent="-1" amplitude="0.1" type="gamma" />
+<feFuncB id="feFuncB75" exponent="-1" amplitude="0.1" type="gamma" />
+</feComponentTransfer>
+<feComposite k4="0" k3="0" k2="0" id="feComposite77" k1="10" operator="arithmetic" result="coefs" in2="invhqminhp" in="hrgbminhp" />
+<!-- 10*(hrgb-hp)*(1/10)*(1/(hq-hp)) = (hrgb-hp)/(hq-hp) -->
+<!-- The following uses "iterative" refinement (or at least something similar) to improve the result, but it cannot cope (well) with negative residuals. -->
+<feComposite k4="0" k3="0" k2="0" id="feComposite79" k1="1" operator="arithmetic" result="hrgbminhpestimate" in2="hqminhp" in="coefs" />
+<!-- (hrgb-hp)/(hq-hp)*(hq-hp) = (hrgb-hp) -->
+<feComposite k1="0" id="feComposite81" k4="1" k3="1" k2="-1" operator="arithmetic" result="hrgbminhpresidual" in2="hrgbminhpestimate" in="hrgbminhp" />
+<feColorMatrix id="feColorMatrix83" values="-1 0 0 0 1 0 -1 0 0 1 0 0 -1 0 1 0 0 0 0 1" type="matrix" result="hrgbminhpresidual" in="hrgbminhpresidual" />
+<feComponentTransfer id="feComponentTransfer85" result="invhqminhp" in="hqminhp">
+<!-- Computes (1/100)*(1/(hq-hp)) -->
+<feFuncR id="feFuncR87" exponent="-1" amplitude="0.01" type="gamma" />
+<feFuncG id="feFuncG89" exponent="-1" amplitude="0.01" type="gamma" />
+<feFuncB id="feFuncB91" exponent="-1" amplitude="0.01" type="gamma" />
+</feComponentTransfer>
+<feComposite k4="0" k3="0" k2="0" id="feComposite93" k1="100" operator="arithmetic" result="coefscorrection" in2="invhqminhp" in="hrgbminhpresidual" />
+<!-- 100*(hrgb-hp)*(1/100)*(1/(hq-hp)) = (hrgb-hp)/(hq-hp) -->
+<feComposite k4="0" k1="0" id="feComposite95" k3="1" k2="1" operator="arithmetic" result="coefs" in2="coefscorrection" in="coefs" />
+<!-- Combine p and q -->
+<feComposite k4="0" k3="0" k2="0" id="feComposite97" k1="1" operator="arithmetic" result="qminpc" in2="coefs" in="qminp" />
+<feComposite k4="0" k1="0" id="feComposite99" k3="1" k2="1" operator="arithmetic" result="color" in2="qminpc" in="p" />
+<!-- This has a slight chance of failing, as alpha gets larger than 1 internally, but it really shouldn't be a problem as the specification clearly says that it operates in premultiplied mode AND the results are clamped to [0,1]. -->
+<!-- Reconstruct original alpha channel -->
+<feColorMatrix id="feColorMatrix101" values="0 0 0 0 1 0 0 0 0 1 0 0 0 0 1 0 0 0 1 0" type="matrix" result="alpha" in="SourceGraphic" />
+<feComposite k4="0" k3="0" k2="0" id="feComposite103" k1="1" operator="arithmetic" in2="alpha" in="color" />
+</filter>
+
+<filter color-interpolation-filters="sRGB" width="1" height="1" y="0" x="0" id="replaceHue">
+<!-- This is mainly to make sure the flood color has the desired effect. It would be better to use "real" colors and compute the hue from them instead, and if so linearRGB might actually have an advantage. -->
+<!-- Set up p, q and q-p -->
+<feColorMatrix id="feColorMatrix7" values="1 0 0 0 0 1 0 0 0 0 1 0 0 0 0 0 0 0 0 1 " type="matrix" result="r" in="SourceGraphic" />
+<feColorMatrix id="feColorMatrix9" values="0 1 0 0 0 0 1 0 0 0 0 1 0 0 0 0 0 0 0 1 " type="matrix" result="g" in="SourceGraphic" />
+<feColorMatrix id="feColorMatrix11" values="0 0 1 0 0 0 0 1 0 0 0 0 1 0 0 0 0 0 0 1 " type="matrix" result="b" in="SourceGraphic" />
+<feBlend id="feBlend13" result="minrg" in2="g" in="r" mode="darken" />
+<feBlend id="feBlend15" result="p" in2="b" in="minrg" mode="darken" />
+<feBlend id="feBlend17" result="maxrg" in2="g" in="r" mode="lighten" />
+<feBlend id="feBlend19" result="q" in2="b" in="maxrg" mode="lighten" />
+<feComponentTransfer id="feComponentTransfer21" result="q2" in="q">
+<!-- q without the red channel -->
+<feFuncR id="feFuncR23" slope="0" type="linear" />
+</feComponentTransfer>
+<feBlend id="feBlend25" result="pq" in2="q2" in="p" mode="lighten" />
+<!-- p in the red channel and q in the rest -->
+<feColorMatrix id="feColorMatrix27" values="-1 1 0 0 0 -1 1 0 0 0 -1 1 0 0 0 0 0 0 0 1 " type="matrix" result="qminp" in="pq" />
+<!-- Set up coefs -->
+<!-- This is what determines the "target" hue. In the ideal case this would use feImage to get the image data from some other object and compute the hue from that. -->
+<feFlood id="feFlood29" result="hk" flood-opacity="1" flood-color="rgb(80%,80%,80%)" />
+<!-- This could also use an arbitrary image whose hue has been determined. -->
+<feComponentTransfer id="feComponentTransfer31" result="coefsq" in="hk">
+<feFuncR id="feFuncR33" tableValues="1 1 0 0 0 1 1" type="table" />
+<feFuncG id="feFuncG35" tableValues="0 1 1 1 0 0 0" type="table" />
+<feFuncB id="feFuncB37" tableValues="0 0 0 1 1 1 0" type="table" />
+</feComponentTransfer>
+<!-- Combine p and q -->
+<feComposite k4="0" k3="0" k2="0" id="feComposite39" k1="1" operator="arithmetic" result="qminpc" in2="coefsq" in="qminp" />
+<feComposite k4="0" k1="0" id="feComposite41" k3="1" k2="1" operator="arithmetic" result="color" in2="qminpc" in="p" />
+<!-- This has a slight chance of failing, as alpha gets larger than 1 internally, but it really shouldn't be a problem as the specification clearly says that it operates in premultiplied mode AND the results are clamped to [0,1]. -->
+<!-- Reconstruct original alpha channel -->
+<feColorMatrix id="feColorMatrix43" values="0 0 0 0 1 0 0 0 0 1 0 0 0 0 1 0 0 0 1 0 " type="matrix" result="alpha" in="SourceGraphic" />
+<feComposite k4="0" k3="0" k2="0" id="feComposite45" k1="1" operator="arithmetic" in2="alpha" in="color" />
+</filter>
+
+</defs>
+
+<g id="layer1" inkscape:groupmode="layer" inkscape:label="Layer 1">
+<image xlink:href="data:image/png;base64,iVBORw0KGgoAAAANSUhEUgAAAJYAAACWCAMAAAAL34HQAAAAAXNSR0IArs4c6QAAAv1QTFRFCAIG
+BwQTAgcFGAUMHQUNFgcZEAgkPwMWPAQjMAokVAIOLg0TIQ8wJg8iKw8bXwQHRQ4OIBcmJRcadAER
+fAIHdQQibwgwNx4PaxACNyEsVBknNyMjNSI6Ths2RyEzPCNMSiQqbxoljxIiNS00NitNWyUhNDAp
+XCcRNDFDYysEjBsyRy87RTEwQTBIOTgaVDAiTDYZSzYnjiBEgSpEqSA4YToPUDtPTTtcVkEPWD0+
+SUBQUD9HUEE9SEkRczdGXz1QjTQkiTQ4O1ASRkViaD9Efzs6eUAbYkU3fkIEVko3fEAvSU1Wd0I8
+qjYfdEoEpTVGSVQrTlBHqTg1bEsvb08EaFIEqTk/WE5eX1EvaExMX05VS1wHXFgEX1BLeVAkb1FD
+TV8eb1YbZFZDblcpXV0jglNYfldaemAUs0lVrE1CdVxVc1xdgFdrYWFrbGBWamBdW2VajF4VbF1+
+nFoRnFJhe15PgGQFq1BTcmJPa2ZPdGsFcmoZmls4lVpSdWJumVxIZ3AZdWhEW2mHjGBUl2YBimNH
+kGkAkGU3ZHBMiG0Ag2o3c281aHMzfXMAk2NnWXKKjWZoim4kXnN9gWtrhGtkiWtccHB5e29kf29c
+a3Rqd3JbeXBstHECuG0frnIgrnYAumlrpnwAjoEapHogmX4dr3RDZ4KXgYcdnndyiH19hH91lYYA
+j3x1iH9toIMAiYoBiYBmk31vln1oeIV4qH07hICIhYdAsndgeYSIm4I6fYSAmIJKr3hxqH5Mo3qH
+kYZdhIeftooFroJxsoJmp4ZkroaAo4x+iJKWi5KOjJOFuY41nJB3oY2GlI+ZmpB+p413mo+Pl5GH
+tJcOrpY2x5oTxZOCxZZ5u5eRs5qRrJulnKCwuZyGvpx5r56WmKSirqCOqKGWrqGHpqGgwp9oo6Ko
+xaFbyawjs6qBqa+fy6uFxKqkv6uwuq+dubGVsLG8uLGms7Kzz7RYq7az3bNg2bdzxMC4vcPJvMW7
+zcWs08iayc+vzNTY0dbN4+jh7u3OwPYOzwAAAAlwSFlzAAAN1wAADdcBQiibeAAAIABJREFUeNpM
+/Atc0nf7Bo5/rFxPB81DrcdDamk9LjXXyUKx1tI505+uZSlOxjLTkgH5LJV0WuahqbkOKJD8f2se
+nhL84CNMkszmPIQaZy3ToQFqZQbCU6GurNfv/tC+r9f/jSCSk8vruu7rvm+KIeNqEYORz2IxmhV6
+o2VhwWKZs8B5Ma4bf2kxm/unxrncl+ZJs3kErrPwyJtZ85x57sWI+Y0hmSySy+W9cnkffPpDLldZ
+74nkvXB/RK4ymfVjOsu7hXfvFhYWTl376adT1347dSoq+ODC2xmzxcwViUXmcZXaPDVlNpsm6aTR
+6O0H5k37tu8LP4C8MFtMFpNpbm7OYgA479/NzcOnuXmvwBVut2+3JXI4IpWc+Sdg+h4wWebgWWZn
+LbNm87hljtsr7gNQAATDNgnX3l75n6pe7CuVfN6gGxvTmh4tYOeNwdH7lPWE+B5cWJiZm60Ui3tF
+ZrP5pdmsejkz0pcfody+fb1leF94+L4DiMU0Pwdgv79s1AJP85fhK4wti3R00SK3BJxZns/kMswm
++O/zZ2fNM4BsZHZ29u3szIx5bo4rkovlvflqmQoOgFHJJ+Uf+JtUvTAaH2nHjGOmhQ/HJfjUQUAW
+FbUpyjI3DjqIe3t7MebNMyOiKfNkH084t1xp2nNgPSl8DcBCWdPzJhQ1AZaFefS5aRpUNC/4+69Y
+sXXrT3jc47lxMYAyj+dbZswLptk5YMo8MgeCgpbjcjFQteeAqysR5UjUGHVwJv+YVKmmTcN6A7Bl
+0r77GxePRvP22RQVFbXw9q1ZMDsn/0P0cnbOPAMqiGbk8n6z6eXMowMnT57cvg+xvMjfO22xpMpA
+QtPctOb93DRoOj/n4rcMWbsWF98RlMzMn8Rw5Ztn5kwvZmdfTGLEj5tfmKwKmNQL0eHLo/dEu7q2
+jkzKR0b+HBkBk5kMJpPMaDRrp416cOz8guV7icJ7ZcCpkIPB8rn8l1TaJOZIMAQoMCs3M2Xwo8yz
+c3uWZmfvw2B5eIC59sruYb+9xTA9bzFNWSzzhwoObSTgCzik+A6metyks5i+H7PMQRUAWRZAk6/G
+GBx/qRuRczn5QezwAxuj2z/A+nNc3ms2TZumAdnY2LTRdHlh4cX8C5NMzfe2dfH1poxwzWQuXiQS
+ExPV85bZl7MzZp2ZgckJKC3r14eHI3M6y4sxsL3BBKDmxsD/ZhOoOL+VzfDy2lqQSKIWMI3mF2bd
+u4X3KhAaPAUkmV6aLVMYhZOTOo56XC7nMsGD7yaBJtWkfLxXbjCatGMGk2FaqzPpQIgXUEgynuIg
+jnvKjmqZm+0Xc0Si3kicF51sxhw7NzmHlfnMzOzI5NjyQOQF+M8yNjJtmje9mQP5NPCzxk2mN2Gq
+Nk5ygr+Ykgg/1Gx+MbKx/R3HAPcwtsCoJoBqHp95aTLrTBj9Ivgjs2lEDmSBNhqL6Z7RYBoDdIYR
+k9EASkD08KQUP5Tjhps17xWJoRDbQvzCKgVmcCvAUVlemOWiP2dnekUiGTJrScx/Y7LQ3hwHKkwm
+BVem9zCNqcbH1W3MrU6Im12QPkxn6J0c2XhWB2hGmCNv3717hxGFWX8SaMNAvRjnYA+9fDH9fuTP
+SRG4YmzMZDJrDYbpKZ3BaDZClcOvwqO5NP+2TG0eF4n3Tprl4nQiQc+YgdqflcuB/xkwvkjHZKpk
+iGXOpJufs3z/fo8FkzKMk8+YMo/oRGP573RG1goHVy8vQhghzItF6lIDIbrjpD3ReXlz5pcW0+Sk
+HIOGPSFmuhfz76cfDTXAoyrDI5NBrx8DUFqTwaCbNhl+t5bjHjdHCcGO95IhEjHNk2LxGb9kvUTU
+1/vHS4jjkclZM0Bi9FPUXMRkuizTWyzoex7klUmWH0aeIu+Vz06q8lHdyLiay98YSEonEUgEOuGM
+fzJHdTw67Gxr1wtrK7DIQc8pwDUy/f69RtnVrRnt6hrp7TMZJGNG2TOjVm98AwAN4DKLakQ2ZmYm
+BrFdVjJQpki8ggM6kogkgeyPP8QiCZMi16nmzHImN98sAlgWCEuZ2fL9m3UghUqn0zEYiUBlvipx
+hCsScds4+fkRuPSIZEJmIJGQToiIjgaydBApL02TOrVp7uXI/DtLl/Uou7o0PV26d+9+/2NMN67V
+amWyYaPMgOWEdjifwENN+Y52NP8wQlsbk+IP0JggQ5VOTolX7R1hithzM7NmlY4z83Jch4z/xmGi
+IN+wTK/jTXK5vPx8hlqm5jDzVSqZiseVq5h7IyLCgK0ILwKBEEiKjj4brX0DPoMkfPTm/fTDrr9P
+95CmS9OuzKcRGDKZ9pFBpx02gIbge/3wsAxFeSjB8rMdy3ZZYjzzsWgvQGMSyGTlyGQfU+jxUi5W
+j0NJzpqm5lQyGcITtfH6RWqdTitr2K6BtvrirWlSpW7j+KlR6Ci6cRUPF0EKI0QTvQgkEinwQDjp
+QHhyWISHq+zRvYfdo92gXU9X12jrkGbo+ehzjYacni8xyUzDOgO4Hq3UToPltQo1H2XzOGqT5cXl
+ZX7xOC83PKVXnk8ikfuhUER71829zdeZvv9eh15mK4VCJdIH8aECk2J9/t27t5gzLS9UIi63jWuZ
+EsGfMjltvF5mWCBpIzUwgeQKIh44EBYBZgvrxigahdvW7i7NQ/3DIeUzzdBQ4t69MiWW8VNgKhZ5
+WGfUGoZlMpSDoijbYpkdXxni7+VPpLr4exHpBKJKPju3MAcMcdUoA76btXdcP2UCWH0q+fjIO+tZ
+wKBZLDpVW1ubSmSZ402qJsf7IcE4bfnEMDouIjAien30geiIiGQSKbqrW/mwa7T7YZdm6JlG3zU0
+JHymeTj0MYEg0QieP3+uGR3SogKjXqscHpNxebw2PoOXbjKJ1BocHu9/ZisO55/OolGYI2qubvz7
+Eex7VEYeiho1HgZEN6WbhLy2YrKeWcuCZZzb1y/mQR6oVKI+eS94jcflwGEyww78C+giJBOScaSu
+ntHRrqGHGrDUkOs6JUg4r3mo2YxLl2ofASz4s+fNEsNYpU4i06EslMtmMzgmE5drMuIDcX7cvbOz
+Y1SSHy5MKBgfl1kYPA6NIyAzpt7d85AhMAVA613Z+2fvDIZrlgvDgRm4ahOJFuZ03BG5B5un6u1V
+q/qYUJbXKG2gYTQughBNInWNDvWMdg8FhhlHn/81r3n+/NHCfFfhvj0Epc7wfOj586HuLmGDUa+T
+STlqXj6Lzc5ncSqnDfnqKYPRy58nIKsN6bhkLdoAdTtiSb3XfpzB4KE8I4QxYraewEA7dxh/xs3j
+XOhrIq6qjSsHEc19KnWiVzR4TzQ+rlOpuOPj48BZfr9anY9zHR3VvHukefjXe0D0YgE065r9a6j1
+9HYvvRHYGn0+2lPFv2wYU2slCh6DxWLT2exKvsqSz50Du/F0KlQx73V5bmFENj4+qbJ0PWwPT6Wg
+HJ7W8mYeMc/CSGxaHrbRdd2hQnV/n1rFG5f3YVOdfGEBxpdJk6lwalymGlfpVH35fTA5qHk8HkBT
+vxl6rnkHFlrnCrBkJqXm4bCRoXyU13BaKzM9x1QcvcsQzGuhxiUqLicxnUVdYPPZJpNlzKDL56Iy
+rXYenPPWPCnTjastyocP70VDuLGeGU3zSLyH+5xpyivMdSPuwFmJWrRuXKeTySBWtTDNy2Qqy/j7
+PJ1Z9VglV/dNjc/N6VTwOMSGCp1+bvwrBJ69ah24e364UDOtVBmHz+ad/lmiHpp4Pjr0fOIuQTIP
+OaFVQ2yxyCjPMsbOJ+t0MLGYVahOppkHVG/N8Ixj4xDJD9u7olMFldBz5hEc0dbGzjEwLMzBNXqP
+QqKWQh1LNFr4XpO1MH8//g/Nixls5MNWATXWmqew9vf+Uevzh/NE4MRVDz0n39Qw3d7113DDz+0a
+o3FfNCYi0NXQI9HLZAoFaEgmkwXoO3LDm7cLC2/BH6hKbbRghWaeHIcxBGC1d3XdK6ziKQ0LAMuV
+bvuRj21VJonerJVAy4dRRG9A3y9MRx8vzMsrzNueLB2bxbr0jNWFlvdw5rsgFlpHR58njz58vnT9
+o6Fu489YUDwHn/2cnZ19goBZHtPx8ryWIZXxQBsyA0X5XIjGtzAzwww/KRNNG7G0nH1p0pnGFtox
+WMquPWrGizEFwiJX7qL47N+oBThGrFcYTXCjiy4sPJuXd5qwNPoAnZMP6fx/iN5Pt997mN36EFL0
+oeY5UaN5aNTCb9n189C8xrANBsufw0+fzl4TNjphxTWkRyFK1fm0xHw+j9FcCTDegnILMMCPq+et
+Af5uwTwybvr9XjvWwe51cbnDJhMCw1r8Vd+rGzVjWr1Bi4GbNgJlhYV7PaApey2LDqeDxXmPrJB+
+Pd1+796v2GltHeoeGtLMNwy1apTK9vbC6I8/Tl2/b1teeHver2uyY/dOTDzvBlgTz3tQLspgcVgM
+AZ/HFzAAkhXXwuwk14RiqN5h7pqXye69a9jD43JZTOAFgS6Rz6yuZuu1oB2MIoZpyzxMu3lnvQIL
+w/ME0qroZA6Hx2tQdq+Znn4BmLIB1L3sIw1YPxyS9ABtmqGuVevz9u2Lzmu919r+656PPz59env0
+6NCPd1t7wGSolM/jUFEeG0Tksz6ICB1udnbEyHj3ga45mUmlqlS/E6QTvfaqdVotYjAMEwQKLYAy
+GfXgd6nWaGmAUMjTG41n86ILwzeSyCwGD/rexlXT2b8++rU9+2Rsd/dnlx/C6AAGu/twaKirK/ps
+XldeQytX2J137/Tx5cdPxu59PtTa3d3dCpVKxlCBs1jsqqp0FBsHMdcDNotOhvWVt28tIzoug4++
+XeBxyCwiwUuCSI06iQnUgxHXIIMJSSAVSKXTpp9b86rA8NFVYQciyBwup721deOq9Seztx/oIPW0
+NnQfkUIzHOp+CCE62v6wu7tnT3FV1543hZe7Tp/O2/7zvtPbRycm/hne+gUUZCUjv5nOZ/HZzc0s
+NsOKCmNsBoRUqbB7b2ffvuDxYfTh/s5CeWQOYwyRmsag/GCO1BplMIvoFBB+MoMxr7hQX3j2gIfb
+8fUkBoeBwYo+EN5cVdyq/+tZV/vZPKNmCOw+hF2G2oe6vUKqen7+XqmU/H46+9f1p09k7wsbHe0+
+dPcL8JeQxeLzeDDXcBgstBLcNWtFNmOxvJ3E6Hq7MPsW7MtV8xhkHj8fDZPpkB4jLCYwbxu0KBkc
+r4PRzaCzWH7O83CR5eX54y+7htHgV2jt6j5wdpVr19ni0fVdzw3D995ruiSjz2HI0jh49bS26gWZ
+VTIduaq99fSvHxO2Q0iA6Sf+0/MtwFImMjg86NOVoBM7n22xEoSRBjmhUr2bfWuZTGfw1GqUw81H
+0csMnn4Y2as2mXRGo0HH57MN0zqtYcw0NmwxFRauC1FHF+LwDFccVkZd3a2tZ6NdlVUajWb6r0dd
+zmHKoaHWUVdQsuev9vZ/HDeil6tcU6Xt7d9nb4/es+1k9rYDEKdQjHoJ0MTitKlZPBaAY7MqP2Ay
+z7w1g+1lU8BcJTwD1qYZvHwU7hoMiN4EQAwGaFQyiRaLrBfTU/AptdDPQ3egmNR8eWMgjc3BLH9A
+mPeXpvuvv9qr8gp79h0fxZgKg7kvVdfenddxXFBVdXxdV0/ex9tPbwxfk33iBJj++USDQKJkkel0
+qocdh8GuquSw6WdUGFFzsNi/nZ01j6jeLXDYHBaUKYsHUxlQxpp+AQGh15lAQ+AMVnMtLPpgfonk
+eLR2ihEdvSc6kBGRT+Zwurqjo12X7hUeL+wZlbZ2tf/zeHv3w26Nx9DDh3vCurq7j0crW2GIUXd3
+LU8vdMjcvCY7++PwHkazEn5hdiW9mrxsGa+aw2LhaJns6oXZ2bldUx75l2HXnOTxGBwgkiUAq/PY
+HBgheEYLBIReN6bVzWMvcpnG7kF/N02bVIafMxdvwnu8KQybGTHTqCxOV1dxcSFboyzMG1UaC7u6
+ndtbu9vBNqOjD4svQ5OVhiu7lA1dDdM/LieHbydvwzoQqlAqLYb5YnI68IVbxkqvZtHs3diVLPXs
+3GzYCnuPaGeJ2dxH5sAgxmBU81gM8B/K5nAMC4hB1qwja8H1gA3kNMHyD3uA7nK0A8n1wLvM+Hyz
+mcbicJStrYVnowsbWruqCqM3Ll23vqtrqOvhs7Duoe6NG9tbR//6q+ueMo9xOXp4+d7w7Njsj9cX
+ohyFab543liMEsiV5OqPT2///R5jmRu/mk1+Oze7i04naMZmZ+YY/HRgkQoXDsrj8Bh8BsPwDjFp
+eVqO1ITFlkEvs0xZX6N8YRkXexSymXPBPOacmUFjMBpaW/NawwsvR693Di8OP3JgjeY/XcohTSJ0
+oHnDo+6H8z+nkgqdCwvzTmbvy84+TUKlw2NzhvCx4nlD2c+Z6fTM5FT2xuXrw7f/3sBoYFfOWtaR
+aXSDgfOYyeHwq6vTM8ksNsrnwQfgMr5DDDqJjqNO1RqgOep0pjnoydMW3f7zl87v11+L5zDjmTNm
+shq9111YRVoVvmfPZ9mb8yA0t3MeQhGO/tX1cKhLOdpVdXzfvjBSduzZBuPPeZcZEqVRqx3WdkUD
+W4aey4TkdCKEFqMqPTN66fLft58+PbKw3JWQatR7sKCFcIAkVjqZCuYSpKNgMx3GlsLI4zOU0A11
+BtmYac5k6r10/vzVS5cuXWUyr3Gioh7PTL5hCbpbwz9btS+vWKjMG73X3v0Zd0hzb7Th+dBDkLGn
+cL3zvuxwmEwKC0+gVajROGw0uE4rusL1xW8M+p+JVCLRnliJklMZKJt1ObXrr+kdG1aFpUuG6Xwq
+g4xy2Cj4JJ2B0hjNLFSBiQh9B4ZHPTQg2ZhWAdUov7H/0tWrGKxL8RQKk3IwfiR/QSXo7j4bLjU+
+Kyzsafirq7ArGn1ouDfaPTo01I2ti3l5rT2FDcfJDeG/GO9pDbKNRjTVYGyIvjc6bTCi5HQq0Y6c
+Cg5jp6MQUNOG4a93bFhKMGioMH9DG2SR+WReOoOTz2Nx0lm8d5iIijHtlE4Cs9aYSa3V3Tj/76uX
+rvoAX+fPR50HWIAtSK5Cu7vzihv+etZd2Jr5l0Cg2RPdPa8ZwqYb7KWHruLCvYV5hc7GidFRHZqq
+p6cb0cwx08frT//aNX0azq8hdmBQxnFGZT6vsnJ++tGRIztWL5+eJlfSCZDW1WSsMUFLZ1HZHD4H
+gzUlgRqUSfWy4bEx1aUb54El8XnPqzcuXaLsj4LNMD4qKn5uBsXmAZnxmQAtFuoLM7UN7Q/nQT/M
+WkPtXdn79lXtezKqeaQVTqBnUg10kqEhfcyAMkgO2du3n3Z2/ifDg0cmo6mMKjKjErrJo9NHYr9e
+s2F1HpppdRVK5LFZrGoOn3qGTeKQ3yFTWoUMfAXO0usuXcWYuuQZf+nG/huXrp7fv59JoZyKioqa
+MaPdXWcPSJVd+ujiQmkVWD66ax5WV/B91x7Svm2fva78uapStu4/dycUrMsGOl0Lq8JyEoPBFgg1
+Sr1G47YsNX2Ns/P6VNj5GfPT7dnbTpzIPnkim0zmkKlUKi9ZUclKrwRoHA6dBbBgftAaGXq1TnQD
+czqc8/t9boCKAOs8pe381f3B8ZSXI8x/V0Wvz3wmLS4U3jNWtv/VPg8D/dDzsJPZ2a2a109cw6TK
+lp4ej4mJ/2nJbKlAoB9WOKxJZVxmSfUSveTevewdm48c2bBh9SpyKko2WNpPbF5z4uRJgMZKp+ZD
+cyOhGFnNrGohg8rikRGtdkwhWKdmQCJgkC5hyUC7dGn/JUB56dqNqzfAYlEq8Y3zN/YzjQ2Cwqqz
+XcPk4+8bJAJNXnb29z++enWA9PpJyxfSwZaeFumTiR4tv1mrl4wZHT52dkbRSg0q0T/SK09sOBK7
+YYPzBuc1DCHDMH869siaEwDqxIk8KIdkMBiPnp6cST+D9XI4yLBCxldsAm4A1t+X85xLl3wu3bjB
+vMFlwqercP3zhvVcFhRmotGFe8L2HdFLZPt++B/h1etXecWv//ckT/qkZRT21ScTMK7JwK4O69eD
+aIIwaUOXViMZ3nAi9siJDV+vBr4qLxtMJ07scMg2Zp8EHdMJLCoZQLEIDA6HQyNT6dAZEa1CKv4D
+EwyDxTx/6d/nL+33AdufvxR+No9wCUizAvrjBtTAeUa089Ll4Z99VtiwBx0Tvn71WvL61avRqtcT
+T85qntwXtt6JPjAhUUqGtWNLw9asWbPP2dlVIhNItMPK7A2xmzfv2LBj9WrQ0Qiwjqx6OG+l6ySZ
+TWbxOZVsMpkKy2QlDdIVRXQ3QCaIzkvWpLpxFQvSq5iCywSp0Xv+uHrj7wMPXVpPyvtMU+gcCPvk
+8UqNYPTV/ypfv35V/NerVy0Typboyn/WFR/4TKEfE+CWOzs7r3Fes8N5DexFAFOzY/OG5R9v2Lzh
+6x2bVzsf0ZwGPDtOHGnPPpF9JDWZSiUto4N0ZGoylc5iVLHZCOYmiE/4uIG5CeMJAMB91x9/bD8L
+lsJoumG9nGcfp2k5hQeURkZDVcOYZOjV6+OvXr16P/rkyd3wNQcORG+O7im+i+LpvOXteXnt7e3r
+Y3fESoYlyl/bT//e/uvq5YbVRzYcAeevyv7A08nY1RuANyohEueWmJxeTSUT6VR6PoQ+AvTAMwMo
+jCfM2Bg1GH8ehdln83h/3OjFYJ4HoOcvCU0yCY8dTheQJev2GISvX79+D0y9evXkifMXP27+7EIh
+W9/Sg/qtCsOYArrWwNR14t6jn9c7rwb2dmz49ciGk9mxJ2I3A65tGLAjq1dnnzySRyCFeRAjSJD3
+HDaVlslqJiPwpH+KMVtfspL0f5JdveFxNu/sjwLZyz/OY/a6dOPPP8WoKl+tZR2oaqarBV2Sqiev
+n7x69fr1kycT938U6vXRPRp2T3EzzyE2NjZML5NopBis7Ngq5WWlTK98GLv6xIbYDZBWJzcAX9uc
+oQX8fvp3wJZNIHgsJdLIfHo6h0iDnKdykPPnz4uxOryEQbmB2eoGRtql8x4//ni2VapAH4PfH/f2
+6/rVaplOpUtlr/9LrxaEpaJVwNZrYGri9fOWH1CB/m5PT0Mz7BBL9yxfdVyr1wj19+7d+70duzHA
+QHjvyIbVJ4Gt7F9BxjWbs52dP16z2nn1kfZH2dnkdA6BRCezqDgykcAh01jIfqvTb3ywEIYHLtB5
+zt/wPNB6Nk893vvnDADSqdX9PL5aN6Ue1wneKHiZDJ5A0P3qyavXTyYaLj9v+UUo7GnIJLvQL/Tw
+nJevco5WyhSXhUJZs1SpVwqllXplVTh46eTJI1i4bz65Y/kJ5/UaiQTV9szfyz59HOUyyenE9GTW
+GRqxmpbOQhAbLEUxkm6Ib/xt7RuQ7UwK888ZsxooUisAlLpf6yftV6m1at34iBq6/OVUafer112w
+dQmkz1s2KgWMfx1KPlRWnO78T/BRtFIqEFYp7ul+16JKIUkiBAgN/zwJgh3J/hW42rH8pIPrM1TS
+oNeM/PrridMMAYebTCYySOQzBHoYjYE4LEZ8IN+tFfeh6M6L+9U6QKBSqxXAk07dBsNHm4hLkMBX
+MvhaPT6u3qdRyiZev9478erJz5eftezcmkA/dmDztm2bN4PXAZYQFSq1VYwG59gq4T2lRNsgUKKX
+nU9kb/750fbT7audV51c7aqUSNQCWClPb7+MvZ5LpaefoRPPEGlEMhHxWuG0yPb8h8HhRm8fJphK
+rWjTqrX9QE0/n8tsY2IvMvM4OEkb7JhAF7A3rtMZBBOvXqdOTDz5nvHsPxcKir8twwOobVZUznuE
+UikqaUahHDcrBWxllVIoZKTuyT6yYU+VVKrnrF7lHLvnkVI51rAnPLqSISNQOWQ+h0zEleOpeACI
+rN26AkEQW6z33bjkuXcKQzUGoqmn1P0iZhu30DmEI277DebaRAWHp1Zp1ePAoPVVzK4nr6tGJyYa
+jJqWljJCkN+y5XCcV0MYrAmXCnjSe1WrsFDd0CBsqFRKhJJ09okjDoxh7TB6b82q5dl7xhouZ95j
+CAQyLZrOSSUT0yNwhGOZRCqVQEQckEhkCeK22Aea4iWxJ4Ib0+nGMVgqNY/LjKfwRap+zGF79+gF
+WqOWwcb+ukonn/zzT1HXkwk2RGnLnTupzI884r8gYaiguiClNjIksstrVmHErVkTqxQK0Kr1aGbm
+8hP/ZGi0yuFKZ+fwNQTZcQYKY56+QYJWn0lNp6an03EJ6XQWu5qFqNUuTvbIiq1uUJOU8/u32i4h
+6DD/qNXgKCZcTlGYXM54IKrXaGGE1Wt5f17i0uQKjoI38WRiXnP/CUHsuXKl1zpvFxfn2M3Q+QDK
+9j1oZkNs7Nf7nD+EqlKi1Gw7smNb7I4Nqdrh4WHnzfvC2PmZEtQoY3zPMOrJZBqeSCdSyYQyOoEA
+8xfydtbCXLx4hX+ym9NHl86vxOPcEFtgCnTCcPF4vH41h/k94ZlWoVEOS7RGnY7npWJq+B1cdHTi
+7sTltviV7ntbDhG24j091mzbvM0Ka/l6AnnVx7GxH6L+4zWx97okR3YcOblj9bYwiVb5u3P08qW8
+vYwGXCYqYcge6QiJdAI1ksxJJxLIBCKNRkNmFxYW5rwW46mR/ms/2uWI88fh19rgdFjNqblclUqN
+9qfjpTKjvkqrUUoM+rF5PS5M3dGRqeJP9OSfX+no6ecV4u2SGImz8whfv2b9ZmuTXpW6ahVu1eYj
+JzcDLOc1H+/DBvrt27fv+XjNo4ZHEqlGtmZ4nZZwKIwhE6CaVAKRcJxAOkPg0AkQXmQ2HeHMzVgW
+FtSBOFo1zs3JHWeH98dHOi4m6MatSqpVYTQ+TaCVSmCyHgMRpQIJX9pM4hMkrPiVKz0Ptfh5+eM9
+3ZMPkTqqipev+mCn9ZWrnFctX7UndjOw9fHHAK1KWFWJCvnCZqmwgJ2XffrIEeiN0MmzY+Fe7JGT
+sUf24DKJDHJqJCEznUpG+pn5phcLbxcIZAaJisMT8JFutEg8fvExCkJzAAAgAElEQVTiRB6NRqUx
+kiOesSRCQQeq11cq2Jk8CSoF40uF95555F/xD/RtcU+IiHAndrF7YFv8bDPk1mbnsOOACjhL/Ocq
+q4ZwZaMKvUBRJb0nUSoLTgKi2JOxsScAElxO7jjyNXzFTk9lsziREVR6JBFh8nmJjLlZ0JJMiCAS
+qfjIyMgILzx+K0JN1yuf6ZWZ0ma9XlAoUepQxQqtEtVrtbCTPNMYcXeD7fEB39oWeK3wCkQKhMpo
+V8xan62qOv5P59UYHOhC2CfAlcnmC5olWgVDoIcB7G+KYr+GKQdgfX3ESheBUElIJxBJDLwbGRkh
+99PaKObZuZmFESqNSPAmEomR9tRIIt6p6ll15TNNcyYPFWoipOwqusdevURRqddqtGNjGuHe4ig8
+zveK7aeIA26xDYIstd+Yt23bhlWMj61zDeap/zsEIdrQrJAppdoqo2ZYmA1kHbGiAh3hFnDBSSeQ
+E6npgbhkQjUZ+b6ZmajmLLx993Z2do4WQiCEJBNxkZsicfjkQimPU6UnsQSXG4SBlUtxUzqODjUY
+jI+M/HVBtMSVdiu9Ix2vZHwZY+tlu8hlyRK3FeiqNasyV1nLb/2a1VZc2FXwDOVkCirZzc2ZUqGw
+oSrbCgOYOgnXEyf/JowEuAgsAp1OJRIQESrKj6fNYa9JL5jNkzgcMZlIOOONw7sRMzU8dnNVsxRF
+ZcOHpDhSqkrOUejJHkRKiJ2Hi4ud0zJbu+SWCxkZy2ydkE1OK5YsZfxzFRvm5c0bQDgrOOtNs0Ao
+gxkNrYTxkU8SKKtOYpCsULBzIvbrr3fAXQIOZhvyXtiAmhMRHvexnMOcs74mvTA7O9OLs/dPTj6Y
+EBmCL1QqaJkyCSqBBhnBT7b/B423LCjEHW8PkJY5+h20t3dZtszRYy2ChBxc5LQCsfXaGLEUe7rN
+R6wxah3n16zpqBKmK6ua+QKpIFPKFwrZ6bHYN30NAoKlrBpa2UsNYeCo6cQQIv4ADaHJ56wv3ltf
+wLdSluhPI25KxuMPpmoUCpJWgo49yw+xC9nkgrA3uhE4HiEu9i5Ojo6Aycne0c7Jzsnbzc7e2w1Z
+ZOv9r/Ue8BQbNq/5/zvhGB6IlUo2n13dzNbypM3bMKaysSrEjA+4dmCfoxOJBCqZSqSGuKQjIy8+
+YAJA716quJUy8zpi8iY3fzwRL5QI9FryUv9mDx7Oa+USO/+DibhEew//EEcA5uXitNI/xH6lk53f
+SlubJTZO3o52fku2EVydV61atTn2623btleGhWOJVQ0TIyog8arpPzWz6TBZVEJb2nESQ7LjyJHN
+Vi0xEfflh9FoibT0ZDqRiMz9/nLmxczbmXFVKqmqsKpYx4zEEd3cbPyIOLKfC+7MErslfrYRXl4u
+9v7L/D3s7Vzcndzc7J2cltjZOa60s19pix2YQRAbWxec68d7cf9ctXyJ6oAgLL2ax853dj4kUDSn
+VzVXS9BmOptdoOgQNLMxu38da3X6jg1rTmDBBTGWXUngkKjkRAIbYOmw1+xTM6uqqgq7CruLi41u
+/v54HIeJx/u7ONp6uzg6uS1Z5uLm7ebiZucGLHpZxbOzt1sG+tk5bvK2tYVwgBsbpPzZpkAGdVW0
+tDJMquCwFXwWZ01EoQAlS9IJDWx2cyadRacLq4TszR/kA642bN585MQRsPzXX8eeJLFYlQwY5TkE
+GkJKZXcLq4qLLxRbTw9UA49FY5DT/fD+NlH2tnZudnZOTnb2TsscHe3c7Je5rXBzdLR1tANYANDW
+ztbdxmaZLeJoiySsWAz47Kj6Q1R2s4BPpVVz+Aw+n82RCGRStl7IJyUTIujNLKnwQwVisFZtsAYr
+UHfiSGx1KpnHYHMi8pkcpLCQhKHRY5gy01MFAhY7k1AtVXOT7Yk2Nt5RAMnxg7Xt7ezsXNzsXdwd
+7UNwiX7geUcMGwUBumzA8EgmpiXOjdPMIxKquXw2n1/JR8mBUr7Gi0QiaAUCakREoFIgFIDFd8RC
+JcZuW+VgkEwfb8BS4shJHiOfz2JdbmaTWEhVVXHhkuJigEVqplc38HFUHkch6+dx8GcOui11snH0
+RRy9neyX2NsDYUCeo523o7sL8RSOaGtnt9LOyd7WxhYzlk3CMQC1CIl0IzTzBJCbEjorvZnGRllV
+qD5QGFGpZDdLIwIPNYP7IbcAFhhq8yqHn10f/UOCdceTsZdRlJHPYKEctBkp7inMXEEkLA2LSKdX
+MprDOGqUw1Tw1OgKhOi3FCZ9G28nWycQzc4eXYYRBsfN3i7Ej0hMDIGHHW0Wg61sFq9FPIAzBPFy
+SedII8jNPDZfyuBn8s6wm1E9+RmLrtBnZtIjSGx+89ZsTDcsrTY7L9/j+r3rFAgJBSnhMFAWp4rB
+YTGQYi+oyEj4jwl0N7afy9RLFU/NQWHnsVtMxW/c6uBgv3hRlK0jGCvfNA6wHEE6u35vJ7uDiUEH
+ExMdP3CFICsKbBYhANDf7QykAJvBaj4jYCvYzyCq6JrqZ0Q2KhRmRpCaISAOWTPLOj1sXr4ns13w
+1xGrwxpkPC4D5fC5KA8hFS8WsNMj8CyaF5GWyKH0zr5Uw0DM49kuPnhwKf5TB4elS2xsoiAF/E20
+ZbaOK23hhmljG2RnuxKOVTg4gYE2GFcI4u9dzRZKq9VzfH5zs5RPJ6RmCvRsQ7pCyhY00wsKSAXp
+EQBh24YNq63biHWtPLAv/PTpwoZ7LFTGYwkYKBcpLF7iT6Sx6OprnJC9ib2UoF75y3xUzZMucYvE
+OxBXuHptdPByRHztFwNTy+wW29va2jnZfhRi2+sL4kIw2H6AswyUXATO93AjZPLSFeNtOgIAq7RP
+ZnsptNi/+GBnSprT6R1seqbX+vXO2JK0fOPGdYSIdGpycmYl1TW60gunG5ahMKpzeci3JLe1uDBg
+J/FaCJ4XJRYze8VzPDWX4+cWgvvHprWBDncOLnVYCoQ5brKztYVscPzIEUhatAvCyg5kswE0NoF0
+K2cYW/hqIY+uE7X19ikYUtwy+5BkxQopIV1I1Q9DEVRL2ZkFdBKAA0DpQF86T60h85sv01exIiOS
+1ybKZKiAiyKHolcgLEE0JzERj0+ELYeZSKH8MZfPR528CoiBuE8dcGXJfiRXByfEJgrIWWwL2tna
+LAKObFeCscBPcIGGuAj5W8RIPpvF728T9/b1tnGI/iEHvfGZRgEVuqFWoqRXgdUyC5olCgkd9i86
+m0UcQ7V6qAVOdPXWrZFe/v4MAQoiZh6yt+UxI2HbwOMpV68ymcAWpVe8V81sptqvWIQ4uN5BqAUb
+jzn8w87GMXixo52j7UoQDzj7yKogFu+LETzmdsC1aNHWRB5B97jvtzYV4BKJ/ez9/YmsBkE6XaJu
+lnbcKujo6DiTUrA7t1zQQc+EAYylpLPOoFAcqahXhNfWrVuHZSpUhhwqXuFGhL2fgA8RX71KaRO3
+ia+KF68UMVleZHDzsTKHsiXEAocy/6UOKxYtsvGGPPgIEH30AZD1YrM1AUP1N11ubFZfW19vv7gP
+6GK6RHok4qUCBalSwkaTy8tvdpSXp5y7lXDuUyntDCn5DCFTq1cUXKRXpwulLA6BiHNrmNbpVEhZ
+uf+SgyFENz9i4kEK86pYLI5nUpbgKL+JtXSvKmo6gVBmSyxfXs7Bd2x0WGGz6BTU5EobW1uwvjXe
+AU+CFY7N37d0Nr1PpFL1ysW9vWJmiB0Oj4ctNawSTHVRm1Jefqw8iT5lniq/2FFdzaYWJF82JieU
+l7PpVcKUn5TYXwvL0LFEpDh6hQ1MLN5U/NVTFAqgEgM2BBdACRKL9DQyKkFR+4MFGzs29d+BrICw
+2HTQFmvMtiGnrExhNCVjeKzEQS1K+X1t/b0fTh/FDVorXi2dS6iWVm9JSUlIKVC8tb4dYGYq4kxK
+QkH5T2HG5JTylJ/KO5q/++6rFAUqkml5RCTz1grECR/iz6bAEbe1tV29RmEigRwmLd7NBVedTKXx
+kGsFyzuotreW36IuddjoiHjb2dMEKGGRzQflIgMRrBoXwQF4dHx/G4go6u+X98lFTAoNH5mIoqLq
+amVzZMIUNgCbsX8h/2KkoyDhKPFiykVSx1cXU86lnMv96lzoN999lXCsYOuWSOTbTK9N/geX4Sin
+KNeuXbsqvmoXQsHTilksBgidSKjSGbTkjQ7L75xZ0rG8g0ftcHBd6kaTVclQW5sPVQhzw6L/C1WH
+JStSf2vrF/X1Wt+j0SfuE4koFPtKsZirWrC+V8KsRpOD1HqT7ndpOZD01VehoZHgtZRz5QXlX33+
+3edfwQd8Qu70k1ZssvdiRlGuMoEuoMzpTC2dTZrQK6RapdLIFMlZ/6icC+Tbd7jeWcZpdtjp4IrX
+qbiUxR9sbrM1GVlkTXgbZOv6wGpy/29tIpBPjr1RBNwl6m071TYH4y/QpJJpLQaTF30Ylei0QmAo
+NCWpPCU0IeVWStLFi1+lSM98/nno59hBDt0+lvGpk39UFOXUb0wxBgvnVj74XH9fr4cFWNDXJpe/
+nJ2Z6bP9qWNjhxurzCH6H17+qIq3GPlgqEBYEW0xsgCbrd1WYr+oDeMKMxZcgTAA1kvcO6Pi6ubm
+p40N8yg+XTtvnLeUnys/l3vu3LnylJSL5bdCL15MKd9apTVu+RzI+hxpWlXj5XbmKpD1m5TJDKZQ
+/oGLbH6q1GieKaUSqZop2vtS3jvz8qXXWgfXDpfqso1fLPXCCVRBVkww9rm5LVmy+IOGi+2cCDym
+WC63IsL4AoDWf65/jTBmMs2PyEx6g0noRnpjlCokR89dPJdUfuviufLc0Ju5Fz/9KeGrlOSI5s+/
+SfoGYA2W3ixY8dup365e43RcDabcunqN6HduELZ8/YSgUiDqk0NpTk7KRV5TLx932IH5A1eQidW0
+vxsz4u++zNHW/u+vbM/g+8S9/XKMK7FVwz5MR3E8wV+pNUj1BsmccgxdRpAIBIJn5QlfAaTy8u9S
+ys+lpJRf/OpMudaLkBka+s3n33yHDGZ863/hzKlrd57V3sqPv/XbIL6srKP+7jN9j7CSzuoXcUSQ
+QHVC6eTLmRnXxZD3Bxxc3MIWW7PAFtnq62tjuxgP7kcWBbrhOH1iEXD1+IOCcEBAeS8eYlLZoIWD
+6gxKgUeETlClNybc+hRsFQqZ8dPFc+cuJoG1tFOpR0PhfP4dciuw9OJtyq1bV8vrk6s7mO61xWWN
+BVLJs9sTQjYnv00kEkE5KYW98rGXM/leK+wDN67ww6+zqoYtPFG+to6brFwttrUP44mgF/Zi79jC
+fNX7B5DV10fLv0bwMpr4YxKhbnp4Dl1BNGh1en1KecTum+cuYrhyL14MvZhScIsHgwtg+nx3KJJT
+P/C/xvr7ndW30tJu7d8VdeXKsZgBZc+EUsrmUUTYOzmAgLoWRQY81bopYCzQwd/NdjHWm5FFZyK9
+7b38/04H22QC5IFVuw/OAoR90BmpQWJcsr5KJlNrhZWmR6y1yaZh2fwzaEIp5UkXUz5POffdudCL
+IGRKiixfJZNFpISGIrk1dTWDt2sHE7LKqJAP3ndqAhvLpMI7yo67zGIpE2NLdKe8pu6JJL9X/ueM
+8dm/lnDqP3jJ1svWJtjRxfvvfFhMa/tNLH7c++djSCwga7xXhDkr5JpYvJf4TCtQSLRGg1ai/ijf
+1CAwjKHQc3LPXQQMoSm3bp0DWKEpP6lUYzCbHQpFahubBorKam9V3z5DaztoS0kvq/k0uudOzx0l
+L/M2zCdiOVNc1nf9SQ+3lzIp7526/8W/PmkCa5WAhivoiK+jk5UqR1iJzohemidBwse9j//mrE0O
+Q8S1NtFW/BhD0lCllQ4bUE/3qXmtRjMvYe1OSEr59FZCaBIW8hfLz1387rvQFDUDVXHHkNs5jRUD
+nbVHa0r++5OLLcXnqc+XS5tapMrB23U8LmgoEnF5gzk1d9CJPlQe1Ntzq+JW00s7+0WfNA4U/QCI
+vBd9CHnbrSGPpybNj/98DJMDxpY16QFVvFjURsQp9bJhY/PwmCTI19H992fzz+YtFm35sYIc6Dgp
+36RcvFh+8RyU4OehW44eUjFkSEVaRVZp0WDJAIfI9DoYRUksc2hMu62821RzS3ibAXUIP/lJS2NN
+vaxOJZbfKbnVlGu4cwd/6H5n54MaqMcPJrOxcZqSPx6fmpzsHZdbqcI+4HoN86YY5y+TGLVhhgaB
+2tPX0/0jT5kRGDMqCj49tzbl1sXPv9vyTeTFi+cgHEDJBJ5ChpSl3SqpiCtpuTJQXnumsRxGjbqK
+ptrSK2V3CpgTmLOArydPazrr79TJ71y5XV6Q0nH7SkdOUw3A6nyC/Ovvdrg4Iip/3Dz5kiPGmk6f
+vO/DEJEoZmJFg4vE3scyZkhv3hsQEOzu4kleZ7AYjYaqmxiW0KTQ8tDPQ6FFJmzZ/a2Az5PJkMHc
+usbDWU9zaq/U1w90NOKEt12bOsuartSxrzHv3BEz29pEgs4H9U09d++I6nOkt6rLSosr67NqBu43
+Pe388lnN4O4lGFleEZQAN2abGADJe/+Uy4HlfvDZb4l9EKdML6JCNjz8jD3MdA/wDvDw8VT7Buyd
+15ie7b4ZmvTdFjA98JRSUMCFSV4hgxuk9lZO7k85jVm3G0s6W2ppA/EdMSVNX5aUsNKZbew7orYO
+MffBwIOSu50tg8315RUVh7mc2wM1pU1PSh8M/O9tS0tnzZfYmki6RvG37wVYmN3/HO/tfQygemlW
+rtrEfmca9AqUzeIGB7kHBPgE0BYC3N19UueNkt2hX6VsKQhN2fJdAp+rFEgEqFomkSmQ3vK0ChAx
+p+ZCTU5jU0d9SWNWXVHdhcxA1m/MtrtoG6f09tMHJU+L6ps6mtJqbyXw6xv/N5Cb1Vh3p6O5ZeDC
+QOngU8TeyUvMvObthr3n1Jqj8slJgNWbL27DeqJY7OKvZZgUmeTgAF93n4O+nhtZbQG7POFY3kSk
+YNmQsDbSSyKQStRqNSpDi0lIeW5OY2NjVn3J4Zqs0sbOpv/VlnQ21d05Sivj3bpTdpspGiht6uxs
+qocgaUrKyrndWNFYUdpY23is+EFnTXHJk8GaFTZOETQx0++3ZcDLY2sjhJocn+wTwbrSa4WFx3co
+XigY+b7uUUCT304eqdcnwDPAx8dznb4g6ZuU5MgVhGZpM68BuBKgdCIBSUrpz6rIyWoqSiv677cV
+jU2Nh6/U514pqn96/7b0TketGB3kXq9p6qwHLEmHk0DFiqbGwZwHFaUDNTVFdwYHmzpv27gRKZS2
+XooXcIW9oZU5CRJOyvsoHBhyxNhok0iUSLR8XnwAHM8gIq6a670/YJePZ4Cv7y7Pc1u++nxLaGSq
+QoYCUwSphO0ViGypflqbk5Z1M6si42hj47HGmoyciqTO+uY7VzoGb5dV8Gru9zTV/fK/uoKcmP8e
+zspNyskBZEcHanLu375S0lhW0zQ4uDjSvhomSGxoAAXFbXinl+bexLZ/i8S9UMr9YqzZo6iWFQIa
++ga7B+FCAoKCMLL2x/vscgerrQ3dsmUrnY+9yUMm4/H51XTk8O7QnIGKik/SauNKs2pLaxqTsHyo
+aKpLqx0cqKltAkkH6ls6n3YOJITurtgNmVxemtQ5kAHKDtx+0vmgZcDJjUZdvBcGyF5riMpFm/Cb
+2sDrbUzr4NUvHu/rJSjZKMM3wC8gyCcg5CC4PioIuMI+fIImfTx9zl3k0PgMLspisVABlCKyKHRt
+aVNOWkVc1s7SxpqY2tvf7s7NvVMeN8Cur6spKRlsyik5mgbJWdNZ09iZlpR0Lq08tyKuvqYGlC3p
+bLzdQvUawHn5E53wvRQK8PNHn9jff1Nbm1jE6cVgQff6s1eFkzSr4wEFplxwQEDQLtAvINg3KsDH
+d8bTx8XTx6cN5eNwPImAqxCwBUYk1MYW8uxwTkxORUZjbtPO2w8Scivqa2+V196ubqqrqSgdHPwl
+7WlnxvWMgcak+lDAdbhid1pRTU3pg6YaLFQvdA48n8hMdvGyxd5DB6NWm4u3G1DVZt2qof3I5ZPK
+QCkmobd3SEBUYlD8Lt+g4ODgoIAA32Af5q5d+z09fYkoH+Wx9DIeqh8bM0qQW7bekbufNiblxDWm
+nSttqk06nOzPzCp9Un/zdm59SU1jXWn9D09rajKeZtU1JT0Y2J0UerSztKSmprak5sGDEoB2vfOB
+5vkMGU+NdPJoo1yDHA1xcRNxxG0fogKbCxukYdogoicIGOAdHL8/KChqf0B8AOSDT1BA/v4g3/0B
+wRQ8Ts3jVym0MGUMG7RIym5Hty0JoXH/TalI2Vpau/vTssPM5oLcnzqarjTdvxnT2Fn3S8b9uqaa
+iqTrnTWd12tiOrIGQNCaktImSIi7yuvX7w/MPTERaLBSetlfa4tnYpMDJHtfHxZi4t6+SZ1ME8EL
+CHCP9/OhxAcTA+J3wYUC9gKPxe8P8PX0gc+JyUT9lHbMYBw2GI165CKyyC0JGmboUaccaNo5t7Ii
+crk7axrLsyClYm7W305r+iEr7X81WVnXm64XPe3s/GSgCDPawIPOB0+bwGIPHoCSRlWNndcZHNHP
+vheUBDS/WUdmbPzqndKjmTjIK0gpn+CokJCooPigqKioYEpAcFBQfEAQZjWKezKr0qiHeUxpMMgk
+AuRW6JYVbklbKw5nfXW4MRdqMPdcblJZ1kBtbdJgU1ZuTUvalYz/X8svDx50Xm8qfVBTVJIBLbq+
+EwzfWVrxGmDVF11//p+W651O9hwiMZFo79KXz+QwxdZOLW573DeplKBMJxoEVkDIyr0HgzwAy/4g
+oCp4l0/wrv27fAL2B+8K9ksnEk1Gvd4o0cPNMHJxyxabrZ8m7HZbG1eRlVs6mFubW3EzJ6O+FqK8
+EyvFtIy7D5qarjc9vT6I0dSZk1XTCU2xBmzVWHrnP0UlA7887WmBecKNgyNcIxPxtnvFTEov1oj+
+gHGiTyeV8iCvooJ9fXxtNwUEeAQRg+ODg6OCg312xUcB2Kj9YDo8PZNuMDwa08IIa9RqkN1LtkAH
+/+qrr775b86tuMba0sHaT6/sTEqLS4ImE9eU03Lsy+tFD4qeQr+uedpZc71kZ8n9B1lPi0qySrNu
+3cyqKS0tGnhwHaTc6MZKJi+pJtITXbz7mPlMa9uR9042C/w93b19fYIDfLwd3T1t3QOCg5nBwZTg
+XfHxu4K9IfWhJoMTOWySXjOs1euHpQajFvn01leLQrcc3BQaGpdUW591szR3d+7RnbW743bGZd2B
+BwZ2x8UV1dflAk9FRTWdvzTFNAJnGTElOTlZWf+pKK0By9cMdD44lOHFodLd3DhEWiTRzq/X6jCY
+BPlS6S6gyh1gBUT6OPr6BwUFUXwh5vdTvOMxyqJCKJ7unjQanaw3aoyFGr1ea9QgVwbr1+6s/3TL
+lrWHb547fA4iPukcdMkrScdqUy40lSTVlcUMlGVk7Wy6DvZ6+gCyvaSz5mlcY1IBhNbTitKcphIs
+vq63XPCqxpNdNvGI+RwqEb/MS3yNKZfLUQGBEhziDW3HOwiPDwihRAFZu6J8gjHnBwdDEQQx4+OD
+gqkcaibUoR5srx/WKpFjtYd255YeDf3mHCTlN9/kQHrl5BQcPpz7yYWSL2I6y8q+/KGuJKuz82k9
+PDlUXn1NSc3A084m8H/nQFbF4ZymJqDr+kSPaCsr8Yy9E4tKc+HSiLCp9VMeyx8TBBwsRHftig/Y
+RYkiRoXs35W4KzhoV3wIZVdAcIjn/nhKPNPXx+dMMlFoHNbohzWY57VIQu7tmJ2/xGTUJsUlhZ5L
+upl1M+fcYFJSVkRWVsXuw0nffvv/xtTGZDUNpO0c6ARiakprm0pLmppKAOHT6yWHP4mrB7z1Ay3S
+Ng4tkb7JqZpIdrMjEmkRmxbbR+LQZnbQLnyQb4B7QBA+ICo+0TM4HsjZHwIpHxy1f7/nfh9KcFCw
+ry+eQpcCHv2w/tmYUa9BjpXdzI35V13dJ01l9TeTknIqdiedq01K2Z2R23Q07tu0gZgfvtwN9t4d
+05TTWDoAaB5kdGZ1Xo8Dr3UO5MQdvo7NEKWv38lUai6VvAREpLk4+dMIHCLe7tNIBp8TTA3x2xXv
+4+Mf4BcVEuwTAoYKCIFYgOSCeSJ+FyXg1C7PSCdHmlSvkUiUj5TDeo0M2flJTcuFL2Jqa8DChyty
+QMncm1dyqr+Mu3D027S6tOtFRWUxO4vSDh+GcC2twVh6UPTgaSk8/mCgtjymZmCgc6Bxoq7z/fu+
+/jaciwsNT/O39adTacRTxEgijxF08GBQMD4gKMQfUjQI5iuXAM9d1jkrGL9rFyWEEg+UeW66FrnW
+YNTLDHqN/pn2kQQ5VnDll19aMj75JDcpruzwt4czfsgYWLskrqKorLT+zp0vD3/xbd3Ow0drf/kl
+rggsVdGY1QQ96OnODJAzt6mxqQTYamx58uT9+3fvVKcS8USeH83F0Y+YyPP2oyTieKcIUHkQ4yGQ
+7vFQjL4wYAXHB+zHCjPY99Quyv6g+PgAfOS1xTyJTKJXajRKrUSiQOrvlv1yKKMoLS7hcExL7uGM
+ppqiDHi2mrqytJKSuKKYmKaY0l+Kv/22riSuqan2h4HG0qedT7OyYmoagd/GivqBzpaBJ/c759+9
+e/+OnojDJUI52uKpNJ7dpmsuWz3z/50YEgVcJQIuKDjfYGAKuk1wsG+Qp6+7OyUeEtV9hR9zE1Gi
+lGj0WnRYJpDoh5GjGWV1gxcy6spKcrN2tpTW/VDa+cPtumO1g18eXftpTkZWxpcZOXFFGSVJh+vu
+l9V/WVLRVDRQFFe7uzQmJysNIhVKMqtp4v37989HO5f4E7cSiEQ3OzyVynPaRHNJ9qfkx4cE4YMS
+9xJ3RYVA24mCYSbAB1B5RgW5wxQY7Aup5eZmjx+TSqTDMlQg1ChkQgmSdeXCjxfuZhRdKK7LrSuq
+v1ia0VmTVdOUG1caU1R7LO5K0xedNTHH0oqyvkmLiSspTUPzVvcAAA7iSURBVCqN6Ww6WgNNICer
+IguWk86Sifv1E4MTdQMl/7O1WXKGSk12oR6knmlzcsq3K4hEiPG0gyGE+I2UKArEql8ANgb6wKTs
+4xskVpDIAYm+nosXL/FLHNPD5qNXSpUC2MqESMyFspILx9L+FbezPqG+vu7KlZacirjOc8SdFRkZ
+Md9mfJJ1IeZoWVZ9TtqxuJjSopja/8Zh43JJ0rkVKXFJSYOlg7AytgzeHrwPprextbWhHaRS8cRI
+As/RneaYcCwhckkIII0ghuA9oOPEB8SHBIO7YJpAM6ksnPvKXZ5uyOLFiyVqpUQqgUrUCyolAhly
+7FBBQVndnYyyY/W3d+aUlV3JHUgafFCxe0XtziKHmM/KAc3duLhjVzJ2H03affhm6c1PHAauD5Tu
+Tst5lpPWWDtQUau/Avbv7HzS8rwIFtlFNv7UxGQqleXmeCaETidFJLjY4cIiKHsh4qPiQ/DYFhYE
+ceWDJO/y9l7p4/ORvdsmx36QUCaVSoQSoVQiUEqRtFvl3xZ8kfavb8vv1FzYXV4fl7Q7Kac0Z0VS
+0idZ0TEZMVk7M27uhFpMKMmIy7kZV1GaFTNQ8/SHmlvNFbthB8pKg4mwoqimvnOw5oktstjWZtEi
+mHBC8Di3TTR/dgG9oCDiU8SLFeGRGB8VHxACrSckyhfyIcB9kY+P+0feK7032bu7Ke/JFDAAKQVK
+OiqUMJTI7fKjxYc+iYs5fOwYKeNQ6OG4krhvcko/zf00NCv3aFpMVnXB0cNZpUVxZfVpVypyyrIO
+N4GjBjpLMp4m/auiIulmxUAW9KWmmrqJNwveyJKkRYsWbUVoRHwiHheyKZN+hk5POPapPxJIT0z0
+T/w37GE+8TCMuq90911pY7PyI+zfLnj7SaRSgaABuBJclgiVSqUQud1Ev1Nw4W7K7rK4srL7oYd3
+1pTsXutQezM3Nyk3Li0trjhmd11GRdat2ty0mIqjtXHg8dpbA6UD5253dCTV5+T2NMEM3dT018TC
+/DsEgeENWYRfbOuEx3kT7aJCDtGP0en05IStizYd9AqKCvaDRhgMbHm6r1zpbmNr47gSWfTRJrUE
+Fcpge20WSBoawPKKZuTCj/Tcrz690HjoduaV64P3c1vqBpouppzLTdm9My63Pm13UVJgxiHs3rmk
+rC++uHn4/8kBtnamgc7PknKaGktvP4PFsvG+ZeH9whs361/72BzcZLNoE0wRXmT/BDomIy2ZlLBp
+sX8AEUcL9gkIgiHHJ2DlSp+VH9mstEXA8A0CkA9VoJIGVrNA2iwUSJGMQwllh1oyEo6WHTpQ1nKl
+4mhZS1rSlbqym4P1MSUxh8tiSnbHXbmTlXvlVm3OlR++LLpyG+IjKebwucM5z3JvldRcqWjKGHwz
+aMH+H1v2CLJoyaJFW9ba2CRTl3D8/SPoyXR6QfKyZmpBZMImFzfPEL/gKHdvENFzpeNKR19fWztk
+EaLlC+AoLjfzpEKFlC3MbBYgF1IKDsXdTjl2rKDs0O4LFzquHE67Un9/sCYnbTCu5pOYtJgv6i4I
+i+O+3F1f9EvGL9e/3F3UlFHxr5qKnN1JSdIcbFAs/f9KuN6YttEz7qNFWsWKWr7kaDgWifY4REqc
+Dz45Jei+nPLnNFmdbCQHYWe2o7MU7sKsKJEpFZq1kBQzpEXQxgdtUQnVqSwgyrLqWFRl3brdwUEL
+tB9mNBiX+VY0rsaKRGCFk/b67pVe598H//w8v/f3PG/e933u3Ts6enVycnDSdP5UndVu/mtZ5RZP
+n+pqYiVWkkS2ts4mAoNZIHt7Y7sbBKOL7zVcPH+2oeG8uUSzOzMDVPThQu5rZX5+NreSWxidhxKC
+JHcW5UVelrN8anlYIJhHi6jUn/4cR3wMhhAJRgIKlueFxLKQ9HVG72Yn78SjnSDXAEH01dzck6lr
+I/8zz3AfHN24Gqprs0LQ6aq3nNQpACsYBMwSg25rnbnVQCTtNY3tl7uuXGl4r+FnF8F4fMtccngI
+sMzHbs6M5RbGlNzDm7lcbh6iZV6mB2NenM++kPNzKYSD5SlyuINJMzSCgfakRcCYuC/hA23yevT+
+7dvRO55ounhtJEFw3SO/v7b4+qjw5uTN0cG/Qh2hpvN175rmcoarIHcTKQJrSYFaa21MDAZJmITP
+Wl0T7vcvN9SbE7T3AK+qlpQLEYBkdHbs5sxsZn40LM4/nIW4fDyVoi24LOND0wJOk6RMp1BPf38S
+iz+im5sJDKMJTPDQSTqezc5N5peFRPR2NDH3qBMjCiPTyel/SycHANbB3rOO0FW3v+M0VPWu1dUD
+Qe7GMMAiifgpezUr4QFKJFncbm1sbXU1+OsvX/mpuRBSNzo7PruQySmZntH58TElI2YeKhVomJgi
+ORnNysHBbCrrweZexOVUmsFRIs7w3d0og8WTQhxYihGwLPby9vJdLBpt8RH8Yj+WuDY1NfemMPLf
+w4PX8Pp+CSbbqVBH/emzVmsrVQW1O0VSDJKkt85f3QdolmnsG2RFr63a+37rxcv1Z8x9TNCsklHC
+8xllbDSWiYjzmZgSZk9OIEeedQTTJH2umMaxrBBNI2JQofN3ibTAhDg6GvfFk0wWo30pT/5efGou
+cT2aTTBEkU9JQ5/LRCca/0Y+ONrxtz3bmyUjXZTbXeW0WpvcFAS5WgNeW5j9Y2W1q1oSyUCfyyaR
+eNDb9q7Vzl62NprLDEsfzM4osQXFPPAdjolsJhBWxEoFks+RJM7BeB+HpJBB3OMX/XRKmMNxGscE
+BgdZPgFyGwHh44m00M9P8sKdZGLyfpSg5Uf84tSb4nD86PBwZVZagp0dXZGqNhdUZWu1kM4QZPcG
+va3QkqEbxyczQYlyB63mCGCDOBiTF9xnFAqyT4zOZEZnZ0bHxsZjvaICfvayx8cVKOuHUxyX8jrg
+6ZSczm5MAwW0ONL0/eFBn+BrifoEwC5hSuD5oWvd/MZQ/AlPC763segQn3qVLBS+eXNU3Prg1MQ7
+20bJLFu2v28WPtxXS/t7Kuv1N/1CNVRjwKiclKjq9/1iLEzZgZLBTdWXr9yQpIdjANTs/MSsqFAT
+oksUKRd7qANrCThXKHKyjHI0k+fwwRciLuHN00jKw2FYnGEYoF0tuODj3u5OpKJJnkZ5TPIJdFFg
+ivvffftq//XRzrrLWr8+s7ulGlsmrP1D44eKaiUqxv5QAG5bMx7o5iZgNydJLOToCcZIb1O9KBUB
+qkCkV3EpYdEV6wmyQda1e1wB1sqn0zA6SAsyCctICHkx0oxyEh73ZDtbHNFLKBiEPNGSHiKSUQzj
+h6b4LP3bTkHopzl+avrpd6939/d3VlasZzPUxAdbqloy4ewZqqpphlHS1pd0cNU2jW3jK+OwUja+
+gsSJX1W5JSBmcCDWp4QzmdlMZEwMj0aooBhxBUnvcaVyrEHJj2jHJYR+PEgyNziYDiIwY2HyOE8T
+OCogQrwfI3jAKVPB0nHiGsMvDvFpYY4YfsXzhW+Ki4WtrcBZqpdqtfVS6ypwXmmnpG5tqTsqeFnT
+dUPVdW1TK2nlAb1i6IfrvbU1lCT1iSwlFYGqf6xk+iJiRuyLBLvYD1kbsNWxrkKPP+Km6XyLA5cv
+kTIymHcMMrTFgW5wKIkmh0NEItGS5rpvpEFG3z3cnQbZDkH0z/FPJpf5wj8L/fIfHvyk7jMwj3d3
+9fZM5EqqUdbN0jKaeQGgNjeB3VStVBowtr8ol806dfW2iTCA1WeGSyVgVzKxcMQVoVjW6/ZbM5WK
+XintQmlxJC4OW9BQWxGuwZthC9xhIQkCQ0My3c2hKEYgdEiAHaiHb8E6mThQVqJlemrq3jxdeD7y
+kBqvdQ70uimKivxpIvf3XEmv6Ia2B7pq1r0xfWhsAj+ufmpsasaAfmjWjvSDQCTF/CAgieORmGgX
+xUDYS1EBt8tpGgs8DrTIpAowPy0HUavzUlyEOdwRIlEGTvcz+Sj3hEi3tLREcQ58RSAdGNc93AmS
+wydg8vqf+3fu9CuKdaaX8lvZLmpgZmJ0YX3r49+VSkZpD3hT0w4BFE1bMssuapq+Cui2OaCXyxVj
+Rww+k1iQ9ARsrkgGjMLGDyk7RbmcFdCMw1Udmmp2kE4cqaE5WfJKTkuRJOE2DEcRziFdoqMJLI/4
+MILj0A4MCRE+YrEQYnjh7j3m3stJbvpm10RPF/Dg5c96el9s3FxbWyktHfxYpMp0oQE4pYELGAGg
+6XpJ3f5C14ECqDm2NxNUgiBBpCIs0GC3y03V1psbXvSTw4oKLXBpL58inPFUjUXqwHkHi9TgYMZi
+sTSjNHE9GxewBILCNIOS3QiHpT1xgmGw5DT/cpKZvznfQwXslL1xxt0bebqQ//YpyDFXPjE3HBn6
+rlkaEphLV9W1H96YgvbltjFQPtQPjdIhKSrjgVY24vWTrXaHzen29wIVOT4BXtcg5iOJxRHEwsQt
+QSpkcfwcETwcjDoxGPakMCEf9fkwBEdxhEgBHWsemRYQrp/nN+Ibg57YoIt1vNPoPutqpfyjhcdj
+X+fWCutqvnSgrRk/VqwEqEAva2vbO8ByJsm+rGjbB3t62ViLzGQabSATA1yn/G6bzXZsnnBQjYp+
+DNXgFsQvOc4NO0kbB/v9IcslGFCpm0RauOj1FIg7fD9Jkp1C5zIIQ7yAIUzn8PRIcsPn9T1DaOrM
+qVrK3tpKLTzN3f/bwl9Wcs+BOpQ+0cB9TVBlwCp1RwOA1O1t02Z/vgW8emvTdHKJHQePY2dZF3Ch
+3+48qZg79lWASodgHA/BDi8ukH2ws8YZCl21oLi3iKYlS3Ne8F3HEgQQLzDXQAgEUI7rAHF75EXh
+ZVYp0hekerf1TGNTI/tg4qmSe/D4N88XHqyurKvmTXeB5wAWbQkQq2SYTiybFFu/ZQDjlW4BPxpG
+2dbV92l7wO60W+vtfQCU/j0Qt8omgMV5ENwZ9p5jSba2puOXVzuutl2lZQ+H0HiKE2ghSRDNBEp0
+/hrlJS4LJIPoJDwIJiQSk3fra63udqq91TwqOCo8Hlj+6/M1eUX7R7m8V9aANmxqP7ZtrWx+KoG+
+aqibmrq5ub2qGsCiS64AS/kd/uraWjPofF8ydK0yMPB/u/CTdQdAY0cAAAAASUVORK5CYII=
+" width="300" height="300" id="image" x="0" y="0" />
+
+<use xlink:href="#image" transform="translate(300,0)" style="filter:url(#replaceHue)"/>
+
+<use xlink:href="#image" transform="translate(600,0)" style="filter:url(#replaceHueFromLayer)"/>
+
+</g>
+</svg>