Analysis of popular API calls
We aggregated data for all API calls, filtering out user-defined functions and other irrelevant data as possible. Results are in data/api_calls.csv and analyzed in data/src/data/api_calls.py.
Columns
api_type: The type of API call—Get, Set, Function, or Construction.this: The object the API is called on, may be empty for static functions.attr: The attribute or method being called, may be empty.appear: How many scripts the API call appears in.appear_interact: How many scripts the API call appears in after interaction started.total: How many times the API call is made.interact: How many times the API call is made after interaction started.%total/total: The percentage out of all API calls in the scripts the API appear in.%interact/interact: The percentage out of all API calls after interaction in the scripts the API appear in.avg%total/script: The average percentage per script.avg%interact/script: The average percentage per script after interaction.%interact/total: The percentage of calls after interaction out of all calls.
Overall, data are tail-heavy, aligning with prior observations:
In [2]: df.describe()
Out[2]:
appear appear_interact total interact %total/total %interact/interact avg%total/script avg%interact/script %interact/total
count 18120.000000 18120.000000 1.812000e+04 1.812000e+04 18120.000000 15662.000000 18120.000000 18120.000000 18120.000000
mean 83.331733 71.206733 6.331973e+03 3.842462e+03 0.927087 0.530690 1.367818 0.515940 39.876583
std 348.429012 279.559604 8.239731e+04 5.046730e+04 6.230344 3.780852 6.853030 2.822503 40.950548
min 1.000000 0.000000 1.000000e+00 0.000000e+00 0.000087 0.000000 0.000088 0.000000 0.000000
25% 4.000000 2.000000 5.000000e+00 0.000000e+00 0.007226 0.000000 0.016281 0.000000 0.000000
50% 7.000000 5.000000 2.300000e+01 5.000000e+00 0.036847 0.008031 0.095799 0.005917 23.913043
75% 35.000000 30.000000 2.650000e+02 8.225000e+01 0.178863 0.092942 0.512494 0.167178 83.026494
max 9957.000000 7181.000000 6.542023e+06 3.267748e+06 100.000000 100.000000 100.000000 100.000000 100.000000
Call distribution
From the CCDF of number of calls per API, we can see that 0.1% of APIs (18) are called over 1,000,000 times, massively outnumbering other APIs. There is also a 2-4 time gap between total API calls and API calls after interaction began, signalling a large number of API calls before interaction (precisely, 45,109,934 before vs 69,625,413 after).
Note: this graph uses the total and interact columns.
To find how many APIs need to be investigated to cover 90% of all API calls, we plot the CDF of the fractions of API calls vs fractions of APIs:
It takes 1.75% (318) APIs to cover 80% of all API calls, and 3.74% (678.0) APIs to cover 90%.
Top 20 API calls overall
As a rough look, we sample the top 20 API calls based on various metrics.
The most called APIs are mainly DOM-related, including many “get” calls on on HTMLDocument or HTMLDivElement and a few functions. The results are similar for calls made after interaction began.
In [3]: df.nlargest(20, "appear")
Out[3]:
api_type this attr appear appear_interact total interact %total/total %interact/interact avg%total/script avg%interact/script %interact/total
12035 Get HTMLDocument createElement 9957 6827 386183 160286 0.383050 0.272003 2.670769 0.975292 41.505193
13516 Get Window location 9423 7181 579071 319716 0.664399 0.615618 6.606676 4.230041 55.211882
2014 Function HTMLDocument createElement 9320 6330 286107 122524 0.296257 0.219127 2.612151 0.967649 42.824538
2211 Get Window addEventListener 7893 6191 122537 31256 0.140030 0.056707 2.405811 0.710843 25.507398
18119 Function Window addEventListener 7166 5668 56988 12036 0.087308 0.031169 1.998809 0.587056 21.120236
241 Get Navigator userAgent 6996 5088 142782 58271 0.173377 0.112728 5.145219 2.053937 40.811167
5125 Get Window document 6890 5156 280870 131302 0.349732 0.245301 2.341958 1.213727 46.748318
9201 Function Window setTimeout 6789 5622 774630 688808 0.847538 1.324614 3.571197 3.203018 88.920904
1990 Get HTMLDocument body 6479 5167 364548 234641 0.378641 0.399859 2.522424 1.552503 64.364912
2370 Get HTMLDocument addEventListener 6238 5288 106708 34235 0.119114 0.059167 1.832010 0.636335 32.082880
15595 Get Window navigator 6201 4558 268025 98575 0.384988 0.216489 4.785670 1.907738 36.778286
2129 Get HTMLDocument documentElement 6129 4566 348638 220531 0.365681 0.365297 3.225033 1.109748 63.255009
14494 Function HTMLDocument addEventListener 5870 5040 38962 10686 0.045599 0.019454 1.434622 0.627379 27.426723
7915 Get Location href 5824 4561 210402 135381 0.291802 0.297886 2.584136 2.093848 64.343970
16137 Get Location search 5601 4460 65008 34421 0.089337 0.081936 2.977207 2.503777 52.948868
14177 Get HTMLDocument getElementById 5600 3731 145667 67289 0.202675 0.151724 6.768832 2.289762 46.193716
2996 Get HTMLDocument cookie 5578 4147 172695 110544 0.324422 0.336975 7.789128 2.641675 64.011118
12711 Function HTMLDocument getElementById 5559 3690 81282 42177 0.114240 0.095513 6.735331 2.271519 51.889717
13966 Get HTMLDocument querySelector 5470 4069 214306 135242 0.261517 0.286668 5.343761 2.736037 63.106959
15619 Function HTMLDocument querySelector 5256 3878 109894 70588 0.143162 0.159666 5.420111 2.722504 64.232806
In [4]: df.nlargest(20, "appear_interact")
Out[4]:
api_type this attr appear appear_interact total interact %total/total %interact/interact avg%total/script avg%interact/script %interact/total
13516 Get Window location 9423 7181 579071 319716 0.664399 0.615618 6.606676 4.230041 55.211882
12035 Get HTMLDocument createElement 9957 6827 386183 160286 0.383050 0.272003 2.670769 0.975292 41.505193
2014 Function HTMLDocument createElement 9320 6330 286107 122524 0.296257 0.219127 2.612151 0.967649 42.824538
2211 Get Window addEventListener 7893 6191 122537 31256 0.140030 0.056707 2.405811 0.710843 25.507398
18119 Function Window addEventListener 7166 5668 56988 12036 0.087308 0.031169 1.998809 0.587056 21.120236
9201 Function Window setTimeout 6789 5622 774630 688808 0.847538 1.324614 3.571197 3.203018 88.920904
2370 Get HTMLDocument addEventListener 6238 5288 106708 34235 0.119114 0.059167 1.832010 0.636335 32.082880
1990 Get HTMLDocument body 6479 5167 364548 234641 0.378641 0.399859 2.522424 1.552503 64.364912
5125 Get Window document 6890 5156 280870 131302 0.349732 0.245301 2.341958 1.213727 46.748318
241 Get Navigator userAgent 6996 5088 142782 58271 0.173377 0.112728 5.145219 2.053937 40.811167
14494 Function HTMLDocument addEventListener 5870 5040 38962 10686 0.045599 0.019454 1.434622 0.627379 27.426723
2129 Get HTMLDocument documentElement 6129 4566 348638 220531 0.365681 0.365297 3.225033 1.109748 63.255009
7915 Get Location href 5824 4561 210402 135381 0.291802 0.297886 2.584136 2.093848 64.343970
15595 Get Window navigator 6201 4558 268025 98575 0.384988 0.216489 4.785670 1.907738 36.778286
16137 Get Location search 5601 4460 65008 34421 0.089337 0.081936 2.977207 2.503777 52.948868
8169 Get Location hostname 4968 4176 136629 68568 0.429844 0.370961 4.515706 3.713880 50.185539
2996 Get HTMLDocument cookie 5578 4147 172695 110544 0.324422 0.336975 7.789128 2.641675 64.011118
13966 Get HTMLDocument querySelector 5470 4069 214306 135242 0.261517 0.286668 5.343761 2.736037 63.106959
7130 Get NodeList length 4664 4065 2382138 1072590 2.861152 1.943079 3.927897 1.756873 45.026359
15619 Function HTMLDocument querySelector 5256 3878 109894 70588 0.143162 0.159666 5.420111 2.722504 64.232806
In [5]: df.nlargest(20, "total")
Out[5]:
api_type this attr appear appear_interact total interact %total/total %interact/interact avg%total/script avg%interact/script %interact/total
10159 Get HTMLDivElement parentNode 2965 2640 6542023 3267748 7.269719 6.034085 3.250519 3.723290 49.950115
10200 Get HTMLDivElement nodeType 2014 1800 4208131 3091512 5.603724 5.771592 3.311296 2.459127 73.465203
12998 Get HTMLBodyElement parentNode 1665 1501 3365475 1217366 4.883114 2.911233 1.474360 1.232955 36.172190
7130 Get NodeList length 4664 4065 2382138 1072590 2.861152 1.943079 3.927897 1.756873 45.026359
12987 Get HTMLDivElement getAttribute 2037 1865 2048964 771237 2.767731 1.546469 2.962922 2.441469 37.640339
17434 Get HTMLDivElement nodeName 1443 1384 1931993 1450429 2.994059 3.124005 4.088068 4.737432 75.074237
1340 Get HTMLDivElement hasAttribute 858 830 1766442 1005093 3.688406 2.609303 2.078833 1.705360 56.899292
14607 Get MutationRecord addedNodes 432 423 1509702 775421 5.358348 4.852424 7.431751 9.234233 51.362521
16368 Get HTMLDivElement contains 542 535 1143362 1053742 2.645657 3.983127 2.107908 6.223987 92.161713
14386 Get MutationRecord type 339 331 1130505 470080 4.330352 3.197764 2.887015 3.495311 41.581417
1488 Get HTMLDivElement parentElement 1652 1584 1119639 646139 1.859494 1.433435 3.000702 2.827354 57.709583
6767 Function HTMLDivElement getAttribute 2039 1867 1117674 508505 1.509619 1.019614 2.401499 2.032517 45.496719
16281 Function HTMLDivElement matches 843 835 1115182 1064906 4.807023 6.144259 2.282807 2.758132 95.491678
7543 Get Performance now 3207 2534 1071387 900360 1.621756 2.036715 4.229508 3.661958 84.036861
17164 Get HTMLDivElement tagName 2146 2058 1063408 740653 1.584709 1.610988 2.923359 2.710166 69.648996
9359 Get HTMLDivElement matches 715 707 1061215 1007261 5.876682 7.905329 2.152317 2.651707 94.915828
4046 Get CustomElementRegistry get 137 89 1051466 1011774 11.754951 12.657662 4.568745 1.805876 96.225080
9514 Get HTMLDivElement querySelector 1871 1647 1030328 988630 1.721028 2.543417 2.069165 1.374590 95.952939
16355 Get HTMLBodyElement nodeType 1094 926 969560 909957 1.680494 2.101329 0.870327 0.668667 93.852572
2513 Get MutationRecord removedNodes 108 107 947656 302503 5.658883 4.521003 5.596039 5.110085 31.921182
In [6]: df.nlargest(20, "interact")
Out[6]:
api_type this attr appear appear_interact total interact %total/total %interact/interact avg%total/script avg%interact/script %interact/total
10159 Get HTMLDivElement parentNode 2965 2640 6542023 3267748 7.269719 6.034085 3.250519 3.723290 49.950115
10200 Get HTMLDivElement nodeType 2014 1800 4208131 3091512 5.603724 5.771592 3.311296 2.459127 73.465203
17434 Get HTMLDivElement nodeName 1443 1384 1931993 1450429 2.994059 3.124005 4.088068 4.737432 75.074237
12998 Get HTMLBodyElement parentNode 1665 1501 3365475 1217366 4.883114 2.911233 1.474360 1.232955 36.172190
7130 Get NodeList length 4664 4065 2382138 1072590 2.861152 1.943079 3.927897 1.756873 45.026359
16281 Function HTMLDivElement matches 843 835 1115182 1064906 4.807023 6.144259 2.282807 2.758132 95.491678
16368 Get HTMLDivElement contains 542 535 1143362 1053742 2.645657 3.983127 2.107908 6.223987 92.161713
4046 Get CustomElementRegistry get 137 89 1051466 1011774 11.754951 12.657662 4.568745 1.805876 96.225080
9359 Get HTMLDivElement matches 715 707 1061215 1007261 5.876682 7.905329 2.152317 2.651707 94.915828
1340 Get HTMLDivElement hasAttribute 858 830 1766442 1005093 3.688406 2.609303 2.078833 1.705360 56.899292
9514 Get HTMLDivElement querySelector 1871 1647 1030328 988630 1.721028 2.543417 2.069165 1.374590 95.952939
16355 Get HTMLBodyElement nodeType 1094 926 969560 909957 1.680494 2.101329 0.870327 0.668667 93.852572
7543 Get Performance now 3207 2534 1071387 900360 1.621756 2.036715 4.229508 3.661958 84.036861
3621 Get TransitionEvent target 326 326 884580 883460 3.460662 5.166294 5.826075 6.594800 99.873386
17442 Get HTMLBodyElement nodeName 1117 1092 905159 878087 1.588063 2.115985 0.662423 0.761367 97.009144
12774 Get HTMLHtmlElement nodeName 777 754 896397 854629 1.931199 2.493193 0.726602 0.848152 95.340457
9483 Function HTMLDivElement querySelector 1871 1647 847881 808260 1.416274 2.079385 2.040353 1.344933 95.327057
3378 Function CustomElementRegistry get 101 71 828723 806370 12.442614 13.151448 5.779473 2.220260 97.302718
14414 Get Event target 2002 1903 824110 783286 1.179062 1.482671 4.650736 5.382586 95.046292
14607 Get MutationRecord addedNodes 432 423 1509702 775421 5.358348 4.852424 7.431751 9.234233 51.362521
From the relatively low %interact/total of some most popular APIs (e.g., 43% for createElement), we can already see a large number of DOM manipulations being done before any interaction.
Unfortunately, the top 20s for %total/total, %interact/interact, avg%total/script and avg%interact/script are mostly user-defined junk. This only shows the custom attributes are highly popular within the scripts that use them.
Top 20 API function calls
Since “get” calls do not usually create side effects, we focus on function calls.
The most popular function calls overall matches our expectations, mainly focusing on querying and interacting with the DOM.
In [7]: df[df["api_type"] == "Function"].nlargest(20, "appear") # type: ignore[reportArgumentType]
Out[7]:
api_type this attr appear appear_interact total interact %total/total %interact/interact avg%total/script avg%interact/script %interact/total
2014 Function HTMLDocument createElement 9320 6330 286107 122524 0.296257 0.219127 2.612151 0.967649 42.824538
18119 Function Window addEventListener 7166 5668 56988 12036 0.087308 0.031169 1.998809 0.587056 21.120236
9201 Function Window setTimeout 6789 5622 774630 688808 0.847538 1.324614 3.571197 3.203018 88.920904
14494 Function HTMLDocument addEventListener 5870 5040 38962 10686 0.045599 0.019454 1.434622 0.627379 27.426723
12711 Function HTMLDocument getElementById 5559 3690 81282 42177 0.114240 0.095513 6.735331 2.271519 51.889717
15619 Function HTMLDocument querySelector 5256 3878 109894 70588 0.143162 0.159666 5.420111 2.722504 64.232806
12778 Function HTMLDocument getElementsByTagName 4775 3038 60070 46347 0.104988 0.118652 2.101945 0.346168 77.154986
5732 Function HTMLDocument querySelectorAll 4647 3765 223991 187374 0.332157 0.412884 1.507670 0.883315 83.652468
2723 Function Storage getItem 4445 3465 80359 53636 0.111631 0.125812 2.384386 1.275023 66.745480
1029 Function Window clearTimeout 4064 3570 122973 97912 0.181246 0.218990 1.177498 2.771284 79.620730
16594 Function Performance now 2764 2340 624855 498159 0.960439 1.144829 3.490532 3.096365 79.723936
8423 Function Window removeEventListener 2715 2396 12394 7457 0.020448 0.018160 0.505042 0.459469 60.166209
16575 Function Navigator get userAgentData 2575 2003 8612 3716 0.019608 0.011080 0.995214 0.463714 43.149094
17624 Function HTMLHeadElement appendChild 2559 1888 17051 7303 0.040659 0.022542 1.667455 0.622105 42.830333
16461 Function HTMLHtmlElement getAttribute 2550 2173 54549 13001 0.088869 0.032554 0.661737 0.118803 23.833617
12704 Function Storage setItem 2530 2151 117494 100433 0.229494 0.285617 1.399357 0.855013 85.479259
7048 Function HTMLDivElement addEventListener 2514 2184 252068 107180 0.391858 0.234839 2.477123 1.046851 42.520272
6007 Function HTMLDivElement appendChild 2328 2052 87416 23624 0.157654 0.065366 2.490404 1.678324 27.024801
16209 Function HTMLDivElement querySelectorAll 2294 2227 574049 442116 0.767773 0.923417 1.696780 1.783361 77.017119
14409 Function DOMTokenList add 2140 1859 75038 50171 0.119709 0.129215 2.026993 1.599572 66.860791
In [8]: df[df["api_type"] == "Function"].nlargest(20, "appear_interact") # type: ignore[reportArgumentType]
Out[8]:
api_type this attr appear appear_interact total interact %total/total %interact/interact avg%total/script avg%interact/script %interact/total
2014 Function HTMLDocument createElement 9320 6330 286107 122524 0.296257 0.219127 2.612151 0.967649 42.824538
18119 Function Window addEventListener 7166 5668 56988 12036 0.087308 0.031169 1.998809 0.587056 21.120236
9201 Function Window setTimeout 6789 5622 774630 688808 0.847538 1.324614 3.571197 3.203018 88.920904
14494 Function HTMLDocument addEventListener 5870 5040 38962 10686 0.045599 0.019454 1.434622 0.627379 27.426723
15619 Function HTMLDocument querySelector 5256 3878 109894 70588 0.143162 0.159666 5.420111 2.722504 64.232806
5732 Function HTMLDocument querySelectorAll 4647 3765 223991 187374 0.332157 0.412884 1.507670 0.883315 83.652468
12711 Function HTMLDocument getElementById 5559 3690 81282 42177 0.114240 0.095513 6.735331 2.271519 51.889717
1029 Function Window clearTimeout 4064 3570 122973 97912 0.181246 0.218990 1.177498 2.771284 79.620730
2723 Function Storage getItem 4445 3465 80359 53636 0.111631 0.125812 2.384386 1.275023 66.745480
12778 Function HTMLDocument getElementsByTagName 4775 3038 60070 46347 0.104988 0.118652 2.101945 0.346168 77.154986
8423 Function Window removeEventListener 2715 2396 12394 7457 0.020448 0.018160 0.505042 0.459469 60.166209
16594 Function Performance now 2764 2340 624855 498159 0.960439 1.144829 3.490532 3.096365 79.723936
16209 Function HTMLDivElement querySelectorAll 2294 2227 574049 442116 0.767773 0.923417 1.696780 1.783361 77.017119
7048 Function HTMLDivElement addEventListener 2514 2184 252068 107180 0.391858 0.234839 2.477123 1.046851 42.520272
16461 Function HTMLHtmlElement getAttribute 2550 2173 54549 13001 0.088869 0.032554 0.661737 0.118803 23.833617
12704 Function Storage setItem 2530 2151 117494 100433 0.229494 0.285617 1.399357 0.855013 85.479259
6007 Function HTMLDivElement appendChild 2328 2052 87416 23624 0.157654 0.065366 2.490404 1.678324 27.024801
16575 Function Navigator get userAgentData 2575 2003 8612 3716 0.019608 0.011080 0.995214 0.463714 43.149094
17624 Function HTMLHeadElement appendChild 2559 1888 17051 7303 0.040659 0.022542 1.667455 0.622105 42.830333
6767 Function HTMLDivElement getAttribute 2039 1867 1117674 508505 1.509619 1.019614 2.401499 2.032517 45.496719
In [9]: df[df["api_type"] == "Function"].nlargest(20, "total") # type: ignore[reportArgumentType]
Out[9]:
api_type this attr appear appear_interact total interact %total/total %interact/interact avg%total/script avg%interact/script %interact/total
6767 Function HTMLDivElement getAttribute 2039 1867 1117674 508505 1.509619 1.019614 2.401499 2.032517 45.496719
16281 Function HTMLDivElement matches 843 835 1115182 1064906 4.807023 6.144259 2.282807 2.758132 95.491678
9483 Function HTMLDivElement querySelector 1871 1647 847881 808260 1.416274 2.079385 2.040353 1.344933 95.327057
3378 Function CustomElementRegistry get 101 71 828723 806370 12.442614 13.151448 5.779473 2.220260 97.302718
9201 Function Window setTimeout 6789 5622 774630 688808 0.847538 1.324614 3.571197 3.203018 88.920904
16594 Function Performance now 2764 2340 624855 498159 0.960439 1.144829 3.490532 3.096365 79.723936
2831 Function HTMLDivElement hasAttribute 858 830 596653 412168 1.246269 1.070484 2.153890 2.032827 69.080018
16209 Function HTMLDivElement querySelectorAll 2294 2227 574049 442116 0.767773 0.923417 1.696780 1.783361 77.017119
11333 Function HTMLDivElement closest 1111 1109 415799 62022 1.477813 0.420875 2.660322 3.038853 14.916342
15923 Function HTMLAnchorElement getAttribute 1630 1553 353941 184154 0.464089 0.381401 1.700309 2.157078 52.029576
4078 Function HTMLDivElement contains 542 535 303535 254118 0.702358 0.960562 1.509068 3.241954 83.719505
2014 Function HTMLDocument createElement 9320 6330 286107 122524 0.296257 0.219127 2.612151 0.967649 42.824538
7048 Function HTMLDivElement addEventListener 2514 2184 252068 107180 0.391858 0.234839 2.477123 1.046851 42.520272
5732 Function HTMLDocument querySelectorAll 4647 3765 223991 187374 0.332157 0.412884 1.507670 0.883315 83.652468
11541 Function DOMTokenList contains 1919 1724 210971 141308 0.421512 0.391151 2.304020 2.127485 66.979822
2490 Function Text contains 51 51 187765 5018 1.569764 0.206079 1.265214 0.071572 2.672490
16718 Function HTMLScriptElement getAttribute 2065 1544 151950 50237 0.378190 0.168485 6.170167 2.245830 33.061533
12692 Function TreeWalker nextNode 139 124 144231 71645 1.507849 0.865876 4.913585 3.978987 49.673787
14823 Function Window getComputedStyle 1697 1384 143326 75800 0.299979 0.220471 1.873741 2.419006 52.886427
4797 Function HTMLDivElement setAttribute 2075 1803 124080 65231 0.200152 0.153838 0.799871 0.543422 52.571728
In [10]: df[df["api_type"] == "Function"].nlargest(20, "interact") # type: ignore[reportArgumentType]
Out[10]:
api_type this attr appear appear_interact total interact %total/total %interact/interact avg%total/script avg%interact/script %interact/total
16281 Function HTMLDivElement matches 843 835 1115182 1064906 4.807023 6.144259 2.282807 2.758132 95.491678
9483 Function HTMLDivElement querySelector 1871 1647 847881 808260 1.416274 2.079385 2.040353 1.344933 95.327057
3378 Function CustomElementRegistry get 101 71 828723 806370 12.442614 13.151448 5.779473 2.220260 97.302718
9201 Function Window setTimeout 6789 5622 774630 688808 0.847538 1.324614 3.571197 3.203018 88.920904
6767 Function HTMLDivElement getAttribute 2039 1867 1117674 508505 1.509619 1.019614 2.401499 2.032517 45.496719
16594 Function Performance now 2764 2340 624855 498159 0.960439 1.144829 3.490532 3.096365 79.723936
16209 Function HTMLDivElement querySelectorAll 2294 2227 574049 442116 0.767773 0.923417 1.696780 1.783361 77.017119
2831 Function HTMLDivElement hasAttribute 858 830 596653 412168 1.246269 1.070484 2.153890 2.032827 69.080018
4078 Function HTMLDivElement contains 542 535 303535 254118 0.702358 0.960562 1.509068 3.241954 83.719505
5732 Function HTMLDocument querySelectorAll 4647 3765 223991 187374 0.332157 0.412884 1.507670 0.883315 83.652468
15923 Function HTMLAnchorElement getAttribute 1630 1553 353941 184154 0.464089 0.381401 1.700309 2.157078 52.029576
11541 Function DOMTokenList contains 1919 1724 210971 141308 0.421512 0.391151 2.304020 2.127485 66.979822
2014 Function HTMLDocument createElement 9320 6330 286107 122524 0.296257 0.219127 2.612151 0.967649 42.824538
595 Function HTMLAnchorElement isSameNode 30 30 117855 117855 8.029145 12.948267 7.101826 11.841353 100.000000
7048 Function HTMLDivElement addEventListener 2514 2184 252068 107180 0.391858 0.234839 2.477123 1.046851 42.520272
14811 Function HTMLInputElement matches 120 120 101906 101899 0.937924 1.010790 1.194562 1.553083 99.993131
12704 Function Storage setItem 2530 2151 117494 100433 0.229494 0.285617 1.399357 0.855013 85.479259
1029 Function Window clearTimeout 4064 3570 122973 97912 0.181246 0.218990 1.177498 2.771284 79.620730
15809 Function HTMLAnchorElement removeAttribute 373 347 99233 96946 1.153120 1.474503 0.726152 0.792383 97.695323
11476 Function DocumentFragment appendChild 555 540 103875 94235 0.567692 0.809485 2.425293 3.739579 90.719615
The top 20 function calls on the other 4 metrics are also less intuitive. They are mostly internal, workers, canvas, etc., and they do not appear much.
In [12]: df[df["api_type"] == "Function"].nlargest(20, "%total/total") # type: ignore[reportArgumentType]
Out[12]:
api_type this attr appear appear_interact total interact %total/total %interact/interact avg%total/script avg%interact/script %interact/total
4048 Function Object query 7 0 7 0 100.000000 NaN 100.000000 0.000000 0.000000
5873 Function DedicatedWorkerGlobalScope importScripts 15 0 15 0 100.000000 NaN 100.000000 0.000000 0.000000
10648 Function KeyboardLayoutMap Iterator next 1 0 50 0 94.339623 NaN 94.339623 0.000000 0.000000
8245 Function Navigator joinAdInterestGroup 22 0 80 0 25.559105 NaN 33.441558 0.000000 0.000000
978 Function HTMLDocument evaluate 5 2 790 0 17.665474 0.000000 17.901950 0.000000 0.000000
14861 Function FontFaceSet add 18 4 18 4 16.666667 16.666667 16.666667 3.703704 22.222222
15873 Function FontFace load 18 4 18 4 16.666667 16.666667 16.666667 3.703704 22.222222
16387 Function HTMLLinkElement insertAdjacentHTML 29 29 29 29 16.666667 16.666667 16.666667 16.666667 100.000000
2636 Function HTMLDivElement isEqualNode 4 4 350 245 14.583333 13.424658 10.373686 10.092331 70.000000
9888 Function HTMLBodyElement get dataset 11 0 11 0 14.285714 NaN 14.285714 0.000000 0.000000
11149 Function Window close 7 0 7 0 14.285714 NaN 14.285714 0.000000 0.000000
3378 Function CustomElementRegistry get 101 71 828723 806370 12.442614 13.151448 5.779473 2.220260 97.302718
6954 Function Window find 1 1 290 266 11.435331 10.901639 11.435331 10.901639 91.724138
7442 Function DedicatedWorkerGlobalScope btoa 5 0 5 0 10.000000 NaN 10.000000 0.000000 0.000000
13750 Function DOMException get name 4 4 5 5 9.433962 9.433962 9.318182 9.318182 100.000000
8856 Function DedicatedWorkerGlobalScope postMessage 53 0 998 0 9.239030 NaN 14.201290 0.000000 0.000000
595 Function HTMLAnchorElement isSameNode 30 30 117855 117855 8.029145 12.948267 7.101826 11.841353 100.000000
3325 Function CanvasRenderingContext2D save 5 5 4700 4252 7.365501 7.344330 8.110019 8.297614 90.468085
4235 Function CanvasRenderingContext2D restore 5 5 4700 4252 7.365501 7.344330 8.110019 8.297614 90.468085
2952 Function HTMLBodyElement getClientRects 12 0 12 0 6.666667 NaN 14.423077 0.000000 0.000000
In [13]: df[df["api_type"] == "Function"].nlargest(20, "%interact/interact") # type: ignore[reportArgumentType]
Out[13]:
api_type this attr appear appear_interact total interact %total/total %interact/interact avg%total/script avg%interact/script %interact/total
4306 Function CanvasRenderingContext2D putImageData 7 7 1378 1146 0.600666 27.501800 7.636882 19.591959 83.164006
2788 Function BroadcastChannel removeEventListener 3 2 3 2 0.603622 25.000000 0.666398 16.666667 66.666667
5443 Function PerformanceObserverEntryList getEntriesByName 4 4 10 10 2.036660 23.255814 2.005391 28.875812 100.000000
14861 Function FontFaceSet add 18 4 18 4 16.666667 16.666667 16.666667 3.703704 22.222222
15873 Function FontFace load 18 4 18 4 16.666667 16.666667 16.666667 3.703704 22.222222
16387 Function HTMLLinkElement insertAdjacentHTML 29 29 29 29 16.666667 16.666667 16.666667 16.666667 100.000000
2636 Function HTMLDivElement isEqualNode 4 4 350 245 14.583333 13.424658 10.373686 10.092331 70.000000
3378 Function CustomElementRegistry get 101 71 828723 806370 12.442614 13.151448 5.779473 2.220260 97.302718
595 Function HTMLAnchorElement isSameNode 30 30 117855 117855 8.029145 12.948267 7.101826 11.841353 100.000000
6954 Function Window find 1 1 290 266 11.435331 10.901639 11.435331 10.901639 91.724138
4237 Function IDBTransaction removeEventListener 13 5 702 21 0.578583 9.589041 3.134371 3.592959 2.991453
1527 Function HTMLCanvasElement dispatchEvent 5 5 5 5 2.659574 9.433962 2.702381 10.952381 100.000000
13750 Function DOMException get name 4 4 5 5 9.433962 9.433962 9.318182 9.318182 100.000000
46 Function HTMLFormElement checkValidity 5 5 35 35 2.800000 9.333333 2.905555 13.967387 100.000000
3325 Function CanvasRenderingContext2D save 5 5 4700 4252 7.365501 7.344330 8.110019 8.297614 90.468085
4235 Function CanvasRenderingContext2D restore 5 5 4700 4252 7.365501 7.344330 8.110019 8.297614 90.468085
1661 Function URL get protocol 90 90 5352 1824 0.524881 6.625740 0.728693 0.645742 34.080717
16281 Function HTMLDivElement matches 843 835 1115182 1064906 4.807023 6.144259 2.282807 2.758132 95.491678
15075 Function MessageEvent stopImmediatePropagation 1 1 1 1 0.564972 5.882353 0.564972 5.882353 100.000000
7786 Function DialogHelperElement querySelector 5 5 540 495 4.846962 5.586277 4.874266 5.632438 91.666667
In [14]: df[df["api_type"] == "Function"].nlargest(20, "avg%total/script") # type: ignore[reportArgumentType]
Out[14]:
api_type this attr appear appear_interact total interact %total/total %interact/interact avg%total/script avg%interact/script %interact/total
4048 Function Object query 7 0 7 0 100.000000 NaN 100.000000 0.000000 0.000000
5873 Function DedicatedWorkerGlobalScope importScripts 15 0 15 0 100.000000 NaN 100.000000 0.000000 0.000000
10648 Function KeyboardLayoutMap Iterator next 1 0 50 0 94.339623 NaN 94.339623 0.000000 0.000000
8245 Function Navigator joinAdInterestGroup 22 0 80 0 25.559105 NaN 33.441558 0.000000 0.000000
6982 Function FontFaceSet load 110 52 735 116 0.429920 2.183735 21.941940 5.717374 15.782313
13606 Function ServiceWorkerGlobalScope fetch 25 0 4307 0 3.182920 NaN 19.438611 0.000000 0.000000
12027 Function ServiceWorkerGlobalScope importScripts 11 0 36 0 0.944386 NaN 18.684029 0.000000 0.000000
978 Function HTMLDocument evaluate 5 2 790 0 17.665474 0.000000 17.901950 0.000000 0.000000
11792 Function Image setAttribute 175 85 628 204 0.101835 0.037405 17.618164 4.888760 32.484076
16387 Function HTMLLinkElement insertAdjacentHTML 29 29 29 29 16.666667 16.666667 16.666667 16.666667 100.000000
14861 Function FontFaceSet add 18 4 18 4 16.666667 16.666667 16.666667 3.703704 22.222222
15873 Function FontFace load 18 4 18 4 16.666667 16.666667 16.666667 3.703704 22.222222
17399 Function PerformanceObserverEntryList getEntriesByType 38 38 3650 3159 4.730062 5.344725 15.360827 19.455392 86.547945
2952 Function HTMLBodyElement getClientRects 12 0 12 0 6.666667 NaN 14.423077 0.000000 0.000000
11149 Function Window close 7 0 7 0 14.285714 NaN 14.285714 0.000000 0.000000
9888 Function HTMLBodyElement get dataset 11 0 11 0 14.285714 NaN 14.285714 0.000000 0.000000
8856 Function DedicatedWorkerGlobalScope postMessage 53 0 998 0 9.239030 NaN 14.201290 0.000000 0.000000
15223 Function TrustedTypePolicy createScript 225 166 5909 3701 0.046810 0.041957 12.566570 7.982096 62.633271
6954 Function Window find 1 1 290 266 11.435331 10.901639 11.435331 10.901639 91.724138
2636 Function HTMLDivElement isEqualNode 4 4 350 245 14.583333 13.424658 10.373686 10.092331 70.000000
In [15]: df[df["api_type"] == "Function"].nlargest(20, "avg%interact/script") # type: ignore[reportArgumentType]
Out[15]:
api_type this attr appear appear_interact total interact %total/total %interact/interact avg%total/script avg%interact/script %interact/total
5443 Function PerformanceObserverEntryList getEntriesByName 4 4 10 10 2.036660 23.255814 2.005391 28.875812 100.000000
4306 Function CanvasRenderingContext2D putImageData 7 7 1378 1146 0.600666 27.501800 7.636882 19.591959 83.164006
17399 Function PerformanceObserverEntryList getEntriesByType 38 38 3650 3159 4.730062 5.344725 15.360827 19.455392 86.547945
3938 Function HTMLAnchorElement removeEventListener 163 163 13043 11362 0.092295 0.108022 5.412695 16.769294 87.111861
2788 Function BroadcastChannel removeEventListener 3 2 3 2 0.603622 25.000000 0.666398 16.666667 66.666667
16387 Function HTMLLinkElement insertAdjacentHTML 29 29 29 29 16.666667 16.666667 16.666667 16.666667 100.000000
46 Function HTMLFormElement checkValidity 5 5 35 35 2.800000 9.333333 2.905555 13.967387 100.000000
595 Function HTMLAnchorElement isSameNode 30 30 117855 117855 8.029145 12.948267 7.101826 11.841353 100.000000
1527 Function HTMLCanvasElement dispatchEvent 5 5 5 5 2.659574 9.433962 2.702381 10.952381 100.000000
6954 Function Window find 1 1 290 266 11.435331 10.901639 11.435331 10.901639 91.724138
1336 Function Window getSelection 103 103 8064 8064 0.517555 0.579214 9.106014 10.571244 100.000000
2636 Function HTMLDivElement isEqualNode 4 4 350 245 14.583333 13.424658 10.373686 10.092331 70.000000
2536 Function HTMLAnchorElement blur 16 16 17 17 0.001361 0.001822 9.398725 9.830911 100.000000
13871 Function Selection toString 95 95 4873 4873 0.350078 0.381274 8.209906 9.385576 100.000000
3469 Function DOMException get message 8 5 15 11 0.007231 0.020454 9.319476 9.318414 73.333333
13750 Function DOMException get name 4 4 5 5 9.433962 9.433962 9.318182 9.318182 100.000000
11893 Function BroadcastChannel close 6 4 8 2 0.045961 0.510204 0.348012 8.333333 25.000000
3325 Function CanvasRenderingContext2D save 5 5 4700 4252 7.365501 7.344330 8.110019 8.297614 90.468085
4235 Function CanvasRenderingContext2D restore 5 5 4700 4252 7.365501 7.344330 8.110019 8.297614 90.468085
15223 Function TrustedTypePolicy createScript 225 166 5909 3701 0.046810 0.041957 12.566570 7.982096 62.633271
Top 20 API Set calls
There are some interesting popular “set” calls as well, despite the many internal functions.
In [16]: df[df["api_type"] == "Set"].nlargest(20, "appear") # type: ignore[reportArgumentType]
Out[16]:
api_type this attr appear appear_interact total interact %total/total %interact/interact avg%total/script avg%interact/script %interact/total
15422 Set HTMLScriptElement src 4145 2406 15143 6913 0.033048 0.019939 2.485560 0.417693 45.651456
11866 Set HTMLScriptElement async 3037 1612 8652 2926 0.023194 0.010426 2.714181 0.277729 33.818770
8172 Set HTMLDocument cookie 2810 2376 25002 15393 0.057419 0.058849 5.065215 0.855618 61.567075
10020 Set HTMLAnchorElement href 2781 2633 43452 29806 0.078498 0.084109 0.583468 0.409084 68.595232
15256 Set HTMLScriptElement onload 2311 1711 15091 7425 0.035689 0.023283 1.407092 0.675300 49.201511
11456 Set CSSStyleDeclaration display 2158 1870 29629 20599 0.057507 0.064276 1.980954 1.334224 69.523102
3257 Set HTMLScriptElement type 1977 1171 4850 2048 0.059195 0.034339 2.318651 0.435811 42.226804
11781 Set HTMLDivElement innerHTML 1938 1811 16250 5709 0.036140 0.021380 0.966634 0.606579 35.132308
299 Set Window dataLayer 1859 1061 1911 405 0.028186 0.011752 9.510403 0.102686 21.193093
7094 Set XMLHttpRequest onreadystatechange 1780 1455 15562 11971 0.039432 0.041042 0.740359 0.471672 76.924560
12681 Set Image src 1693 1131 6466 3467 0.016076 0.011726 4.572729 1.848441 53.618930
13778 Set HTMLScriptElement onerror 1586 1291 12757 5854 0.033671 0.019983 0.744364 0.615878 45.888532
2347 Set XMLHttpRequest withCredentials 1564 1254 8258 6551 0.019977 0.023260 0.648240 0.361858 79.329135
14550 Set CSSStyleDeclaration width 1292 1097 19640 8521 0.065403 0.042337 1.446305 0.382616 43.385947
17552 Set HTMLImageElement src 1202 837 15598 5529 0.059533 0.037416 3.087841 1.070676 35.446852
258 Set CSSStyleDeclaration position 1192 969 21109 2744 0.052430 0.009619 0.507048 0.181102 12.999195
2148 Set HTMLIFrameElement src 1106 838 1383 543 0.011561 0.008270 0.710719 0.609099 39.262473
8985 Set HTMLDivElement id 1084 916 4409 709 0.013589 0.003192 0.643126 0.349976 16.080744
17444 Set MessagePort onmessage 1043 886 1228 133 0.003000 0.000449 0.391508 0.161973 10.830619
13010 Set CSSStyleDeclaration height 1003 833 29251 24464 0.069698 0.081981 0.304656 0.070588 83.634748
In [17]: df[df["api_type"] == "Set"].nlargest(20, "appear_interact") # type: ignore[reportArgumentType]
Out[17]:
api_type this attr appear appear_interact total interact %total/total %interact/interact avg%total/script avg%interact/script %interact/total
10020 Set HTMLAnchorElement href 2781 2633 43452 29806 0.078498 0.084109 0.583468 0.409084 68.595232
15422 Set HTMLScriptElement src 4145 2406 15143 6913 0.033048 0.019939 2.485560 0.417693 45.651456
8172 Set HTMLDocument cookie 2810 2376 25002 15393 0.057419 0.058849 5.065215 0.855618 61.567075
11456 Set CSSStyleDeclaration display 2158 1870 29629 20599 0.057507 0.064276 1.980954 1.334224 69.523102
11781 Set HTMLDivElement innerHTML 1938 1811 16250 5709 0.036140 0.021380 0.966634 0.606579 35.132308
15256 Set HTMLScriptElement onload 2311 1711 15091 7425 0.035689 0.023283 1.407092 0.675300 49.201511
11866 Set HTMLScriptElement async 3037 1612 8652 2926 0.023194 0.010426 2.714181 0.277729 33.818770
7094 Set XMLHttpRequest onreadystatechange 1780 1455 15562 11971 0.039432 0.041042 0.740359 0.471672 76.924560
13778 Set HTMLScriptElement onerror 1586 1291 12757 5854 0.033671 0.019983 0.744364 0.615878 45.888532
2347 Set XMLHttpRequest withCredentials 1564 1254 8258 6551 0.019977 0.023260 0.648240 0.361858 79.329135
3257 Set HTMLScriptElement type 1977 1171 4850 2048 0.059195 0.034339 2.318651 0.435811 42.226804
12681 Set Image src 1693 1131 6466 3467 0.016076 0.011726 4.572729 1.848441 53.618930
14550 Set CSSStyleDeclaration width 1292 1097 19640 8521 0.065403 0.042337 1.446305 0.382616 43.385947
299 Set Window dataLayer 1859 1061 1911 405 0.028186 0.011752 9.510403 0.102686 21.193093
258 Set CSSStyleDeclaration position 1192 969 21109 2744 0.052430 0.009619 0.507048 0.181102 12.999195
8985 Set HTMLDivElement id 1084 916 4409 709 0.013589 0.003192 0.643126 0.349976 16.080744
17444 Set MessagePort onmessage 1043 886 1228 133 0.003000 0.000449 0.391508 0.161973 10.830619
2148 Set HTMLIFrameElement src 1106 838 1383 543 0.011561 0.008270 0.710719 0.609099 39.262473
17552 Set HTMLImageElement src 1202 837 15598 5529 0.059533 0.037416 3.087841 1.070676 35.446852
13010 Set CSSStyleDeclaration height 1003 833 29251 24464 0.069698 0.081981 0.304656 0.070588 83.634748
In [18]: df[df["api_type"] == "Set"].nlargest(20, "total") # type: ignore[reportArgumentType]
Out[18]:
api_type this attr appear appear_interact total interact %total/total %interact/interact avg%total/script avg%interact/script %interact/total
9493 Set Event flow 121 121 72147 57173 0.262724 0.259411 0.292955 0.308683 79.245152
10020 Set HTMLAnchorElement href 2781 2633 43452 29806 0.078498 0.084109 0.583468 0.409084 68.595232
2968 Set Text textContent 42 41 37094 23641 1.142186 0.933173 0.917371 0.731892 63.732679
11456 Set CSSStyleDeclaration display 2158 1870 29629 20599 0.057507 0.064276 1.980954 1.334224 69.523102
9902 Set HTMLSpanElement textContent 497 463 29552 8465 0.262036 0.134479 1.362292 0.089505 28.644423
13010 Set CSSStyleDeclaration height 1003 833 29251 24464 0.069698 0.081981 0.304656 0.070588 83.634748
8172 Set HTMLDocument cookie 2810 2376 25002 15393 0.057419 0.058849 5.065215 0.855618 61.567075
4044 Set HTMLImageElement onload 510 440 21372 11883 0.361391 0.249478 0.810341 0.577726 55.600786
258 Set CSSStyleDeclaration position 1192 969 21109 2744 0.052430 0.009619 0.507048 0.181102 12.999195
12959 Set CSSStyleDeclaration left 565 532 20098 2004 0.078073 0.011633 0.940909 0.591199 9.971141
14550 Set CSSStyleDeclaration width 1292 1097 19640 8521 0.065403 0.042337 1.446305 0.382616 43.385947
16491 Set CSSStyleDeclaration fontFamily 143 116 16992 591 0.161403 0.007508 4.393393 0.188850 3.478107
13647 Set CSSStyleDeclaration transform 704 696 16935 14308 0.090705 0.128078 0.874879 0.462369 84.487747
11781 Set HTMLDivElement innerHTML 1938 1811 16250 5709 0.036140 0.021380 0.966634 0.606579 35.132308
5856 Set HTMLDivElement className 881 726 15672 4497 0.087585 0.033973 1.528997 1.046062 28.694487
131 Set Event isDefaultPrevented 191 191 15620 15617 0.351237 2.103407 0.222507 7.115688 99.980794
17552 Set HTMLImageElement src 1202 837 15598 5529 0.059533 0.037416 3.087841 1.070676 35.446852
7094 Set XMLHttpRequest onreadystatechange 1780 1455 15562 11971 0.039432 0.041042 0.740359 0.471672 76.924560
15422 Set HTMLScriptElement src 4145 2406 15143 6913 0.033048 0.019939 2.485560 0.417693 45.651456
15256 Set HTMLScriptElement onload 2311 1711 15091 7425 0.035689 0.023283 1.407092 0.675300 49.201511
In [19]: df[df["api_type"] == "Set"].nlargest(20, "interact") # type: ignore[reportArgumentType]
Out[19]:
api_type this attr appear appear_interact total interact %total/total %interact/interact avg%total/script avg%interact/script %interact/total
9493 Set Event flow 121 121 72147 57173 0.262724 0.259411 0.292955 0.308683 79.245152
10020 Set HTMLAnchorElement href 2781 2633 43452 29806 0.078498 0.084109 0.583468 0.409084 68.595232
13010 Set CSSStyleDeclaration height 1003 833 29251 24464 0.069698 0.081981 0.304656 0.070588 83.634748
2968 Set Text textContent 42 41 37094 23641 1.142186 0.933173 0.917371 0.731892 63.732679
11456 Set CSSStyleDeclaration display 2158 1870 29629 20599 0.057507 0.064276 1.980954 1.334224 69.523102
131 Set Event isDefaultPrevented 191 191 15620 15617 0.351237 2.103407 0.222507 7.115688 99.980794
8172 Set HTMLDocument cookie 2810 2376 25002 15393 0.057419 0.058849 5.065215 0.855618 61.567075
13647 Set CSSStyleDeclaration transform 704 696 16935 14308 0.090705 0.128078 0.874879 0.462369 84.487747
7094 Set XMLHttpRequest onreadystatechange 1780 1455 15562 11971 0.039432 0.041042 0.740359 0.471672 76.924560
4044 Set HTMLImageElement onload 510 440 21372 11883 0.361391 0.249478 0.810341 0.577726 55.600786
204 Set Event isPropagationStopped 50 50 11723 11723 1.076866 1.692896 0.501635 3.173913 100.000000
11747 Set Event nativeEvent 50 50 11723 11723 1.076866 1.692896 0.501635 3.173913 100.000000
15818 Set Event persist 50 50 11723 11723 1.076866 1.692896 0.501635 3.173913 100.000000
14550 Set CSSStyleDeclaration width 1292 1097 19640 8521 0.065403 0.042337 1.446305 0.382616 43.385947
9902 Set HTMLSpanElement textContent 497 463 29552 8465 0.262036 0.134479 1.362292 0.089505 28.644423
15256 Set HTMLScriptElement onload 2311 1711 15091 7425 0.035689 0.023283 1.407092 0.675300 49.201511
15422 Set HTMLScriptElement src 4145 2406 15143 6913 0.033048 0.019939 2.485560 0.417693 45.651456
5704 Set CSSStyleDeclaration opacity 212 192 7607 6789 0.047285 0.070397 0.306024 0.537926 89.246746
935 Set HTMLAnchorElement onclick 287 274 13304 6744 0.126817 0.105791 0.135959 0.098332 50.691521
2347 Set XMLHttpRequest withCredentials 1564 1254 8258 6551 0.019977 0.023260 0.648240 0.361858 79.329135
Specific API of interest
Some APIs clearly indicate developers’ intent. We look at them case by case.
addEventListener on various elements all clearly indicate frontend processing, regardless of whether they are called after interaction started.
In [22]: df[(df["attr"] == "addEventListener") & (df["appear"] > 100) & (df["api_type"] == "Function")]
Out[22]:
api_type this attr appear appear_interact total interact %total/total %interact/interact avg%total/script avg%interact/script %interact/total
170 Function HTMLHtmlElement addEventListener 435 403 10386 2790 0.042678 0.015690 1.014821 0.156781 26.863085
482 Function HTMLLinkElement addEventListener 357 203 6291 1129 0.165954 0.097912 1.344452 0.234178 17.946272
926 Function HTMLUListElement addEventListener 318 302 1973 73 0.020669 0.000985 0.300937 0.006706 3.699949
1271 Function HTMLImageElement addEventListener 585 539 25288 11160 0.128254 0.094256 0.826761 0.479900 44.131604
1593 Function HTMLSpanElement addEventListener 346 333 7552 270 0.049295 0.002749 0.135847 0.027511 3.575212
3005 Function HTMLParagraphElement addEventListener 101 101 2233 2 0.130516 0.000176 0.138290 0.000166 0.089566
3031 Function HTMLSelectElement addEventListener 236 217 1468 21 0.046463 0.001421 0.465725 0.017949 1.430518
3308 Function NetworkInformation addEventListener 106 103 119 18 0.021710 0.008249 0.167541 0.024107 15.126050
3772 Function HTMLElement addEventListener 750 739 32724 2545 0.183205 0.022435 1.564308 1.332341 7.777167
5077 Function MediaQueryList addEventListener 708 671 5206 2308 0.014862 0.008460 0.791159 0.285986 44.333461
5616 Function HTMLAnchorElement addEventListener 1346 1174 77789 18943 0.184291 0.063770 6.796528 0.760201 24.351772
7048 Function HTMLDivElement addEventListener 2514 2184 252068 107180 0.391858 0.234839 2.477123 1.046851 42.520272
7638 Function HTMLVideoElement addEventListener 233 193 13401 2890 0.273275 0.080812 2.216936 0.073431 21.565555
8196 Function AbortSignal addEventListener 104 104 8336 8160 0.026733 0.031255 0.176075 0.404642 97.888676
9282 Function HTMLButtonElement addEventListener 1475 1215 21572 3449 0.048451 0.011562 1.259833 0.811528 15.988318
11331 Function HTMLFormElement addEventListener 384 302 1962 97 0.050651 0.005543 4.696687 0.949741 4.943935
13049 Function XMLHttpRequest addEventListener 113 97 1176 715 0.010391 0.019337 0.309974 0.138416 60.799320
13077 Function HTMLScriptElement addEventListener 523 481 2340 1130 0.010101 0.013179 1.327969 0.154031 48.290598
13558 Function HTMLLIElement addEventListener 390 376 9582 737 0.096313 0.010782 0.262191 0.054946 7.691505
14483 Function HTMLBodyElement addEventListener 1021 964 10257 4482 0.027587 0.016387 0.603541 0.057784 43.696987
14494 Function HTMLDocument addEventListener 5870 5040 38962 10686 0.045599 0.019454 1.434622 0.627379 27.426723
16637 Function HTMLInputElement addEventListener 623 585 4813 1395 0.014064 0.005429 0.332024 0.059557 28.984002
18119 Function Window addEventListener 7166 5668 56988 12036 0.087308 0.031169 1.998809 0.587056 21.120236
HTMLDocument.createElement indicates either UX enhancement or DOM element generation. Since only 43% are called after interaction started, more than half of them are clearly DOM element generation.
In [24]: df[(df["attr"] == "createElement") & (df["api_type"] == "Function")]
Out[24]:
api_type this attr appear appear_interact total interact %total/total %interact/interact avg%total/script avg%interact/script %interact/total
2014 Function HTMLDocument createElement 9320 6330 286107 122524 0.296257 0.219127 2.612151 0.967649 42.824538
Window.requestAnimationFrame is a clear indication of UX enhancement because it is strictly for animation. Much more of it is called after interaction started. Though calling it before interaction might make sense because the page might be animating before the user interacts.
In [26]: df[(df["attr"] == "requestAnimationFrame") & (df["api_type"] == "Function")]
Out[26]:
api_type this attr appear appear_interact total interact %total/total %interact/interact avg%total/script avg%interact/script %interact/total
9128 Function Window requestAnimationFrame 1655 1422 42244 28852 0.092087 0.097062 1.884883 3.626479 68.298457
Navigator.sendBeacon is a clear indication of extensional features (tracking). It seems to be called more after interaction started.
In [27]: df[(df["attr"] == "sendBeacon") & (df["api_type"] == "Function")]
Out[27]:
api_type this attr appear appear_interact total interact %total/total %interact/interact avg%total/script avg%interact/script %interact/total
1989 Function Navigator sendBeacon 1132 1085 4685 3613 0.011954 0.012742 0.395376 3.096326 77.118463