From 30ff770c930c0146f0318ca90ac949e952ab93c5 Mon Sep 17 00:00:00 2001
From: "Matthijs S. Berends" Note: values on this page will change with every website update since they are based on randomly created values and the page was written in RMarkdown. However, the methodology remains unchanged. This page was generated on 11 February 2019. Note: values on this page will change with every website update since they are based on randomly created values and the page was written in RMarkdown. However, the methodology remains unchanged. This page was generated on 12 February 2019. Using the Using the Using the Using the The resulting data set contains 5,000 blood culture isolates. With the Now, let’s start the cleaning and the analysis! Use the frequency table function So, we can draw at least two conclusions immediately. From a data scientist perspective, the data looks clean: only values The data is already quite clean, but we still need to transform some variables. The The data is already quite clean, but we still need to transform some variables. The We also want to transform the antibiotics, because in real life data we don’t know if they are really clean. The We also want to transform the antibiotics, because in real life data we don’t know if they are really clean. The Finally, we will apply EUCAST rules on our antimicrobial results. In Europe, most medical microbiological laboratories already apply these rules. Our package features their latest insights on intrinsic resistance and exceptional phenotypes. Moreover, the Because the amoxicillin (column Now that we have the microbial ID, we can add some taxonomic properties: This So only 28.2% is suitable for resistance analysis! We can now filter on it with the So only 28.3% is suitable for resistance analysis! We can now filter on it with the For future use, the above two syntaxes can be shortened with the Only 2 isolates are marked as ‘first’ according to CLSI guideline. But when reviewing the antibiogram, it is obvious that some isolates are absolutely different strains and should be included too. This is why we weigh isolates, based on their antibiogram. The Only 1 isolates are marked as ‘first’ according to CLSI guideline. But when reviewing the antibiogram, it is obvious that some isolates are absolutely different strains and should be included too. This is why we weigh isolates, based on their antibiogram. The If a column exists with a name like ‘key(…)ab’ the Instead of 2, now 9 isolates are flagged. In total, 78.9% of all isolates are marked ‘first weighted’ - 50.7% more than when using the CLSI guideline. In real life, this novel algorithm will yield 5-10% more isolates than the classic CLSI guideline. Instead of 1, now 8 isolates are flagged. In total, 78.9% of all isolates are marked ‘first weighted’ - 50.7% more than when using the CLSI guideline. In real life, this novel algorithm will yield 5-10% more isolates than the classic CLSI guideline. As with So we end up with 15,789 isolates for analysis. So we end up with 15,783 isolates for analysis. We can remove unneeded columns: Now our data looks like: Time for the analysis! Or can be used like the Frequency table of Frequency table of Columns: 2 Shortest: 16How to conduct AMR analysis
Matthijs S. Berends
- 11 February 2019
+ 12 February 2019
AMR.Rmd
Introduction
@@ -210,21 +210,21 @@
-
2019-02-11
+2019-02-12
abcd
Escherichia coli
S
S
-
2019-02-11
+2019-02-12
abcd
Escherichia coli
S
R
-
2019-02-11
+2019-02-12
efgh
Escherichia coli
R
@@ -285,25 +285,25 @@
Put everything together
-sample()
function, we can randomly select items from all objects we defined earlier. To let our fake data reflect reality a bit, we will also approximately define the probabilities of bacteria and the antibiotic results with the prob
parameter.sample()
function, we can randomly select items from all objects we defined earlier. To let our fake data reflect reality a bit, we will also approximately define the probabilities of bacteria and the antibiotic results with the prob
parameter.sample_size <- 20000
-data <- data.frame(date = sample(dates, size = sample_size, replace = TRUE),
- patient_id = sample(patients, size = sample_size, replace = TRUE),
- hospital = sample(hospitals, size = sample_size, replace = TRUE,
+data <- data.frame(date = sample(dates, size = sample_size, replace = TRUE),
+ patient_id = sample(patients, size = sample_size, replace = TRUE),
+ hospital = sample(hospitals, size = sample_size, replace = TRUE,
prob = c(0.30, 0.35, 0.15, 0.20)),
- bacteria = sample(bacteria, size = sample_size, replace = TRUE,
+ bacteria = sample(bacteria, size = sample_size, replace = TRUE,
prob = c(0.50, 0.25, 0.15, 0.10)),
- amox = sample(ab_interpretations, size = sample_size, replace = TRUE,
+ amox = sample(ab_interpretations, size = sample_size, replace = TRUE,
prob = c(0.60, 0.05, 0.35)),
- amcl = sample(ab_interpretations, size = sample_size, replace = TRUE,
+ amcl = sample(ab_interpretations, size = sample_size, replace = TRUE,
prob = c(0.75, 0.10, 0.15)),
- cipr = sample(ab_interpretations, size = sample_size, replace = TRUE,
+ cipr = sample(ab_interpretations, size = sample_size, replace = TRUE,
prob = c(0.80, 0.00, 0.20)),
- gent = sample(ab_interpretations, size = sample_size, replace = TRUE,
+ gent = sample(ab_interpretations, size = sample_size, replace = TRUE,
prob = c(0.92, 0.00, 0.08))
)
left_join()
function from the dplyr
package, we can ‘map’ the gender to the patient ID using the patients_table
object we created earlier:left_join()
function from the dplyr
package, we can ‘map’ the gender to the patient ID using the patients_table
object we created earlier:head()
function we can preview the first 6 values of this data set:
@@ -320,64 +320,31 @@
-
-2013-12-28
-M2
-Hospital B
-Streptococcus pneumoniae
-R
-S
-S
-S
-M
-
-
-2017-03-17
-Z5
+2012-04-28
+X6
Hospital B
Escherichia coli
R
S
-S
+R
S
F
-
2011-05-13
-G5
-Hospital B
-Streptococcus pneumoniae
-S
-S
-S
-S
-M
-
-
-2011-03-06
-X3
-Hospital D
-Escherichia coli
-S
-S
-S
-S
-F
-
-
-2016-11-01
-C9
-Hospital D
-Escherichia coli
-R
-S
-R
-S
-M
-
-
+2010-12-16
-E2
+2012-12-23
+Z1
Hospital A
+Escherichia coli
+I
+S
+R
+S
+F
+
+
+2011-05-27
+G10
+Hospital B
Streptococcus pneumoniae
S
S
@@ -385,6 +352,39 @@
S
M
+
+2012-08-19
+Q1
+Hospital D
+Escherichia coli
+S
+R
+S
+S
+F
+
+
+2016-05-06
+Y2
+Hospital D
+Klebsiella pneumoniae
+S
+S
+S
+S
+F
+
+
2010-08-27
+I8
+Hospital A
+Escherichia coli
+R
+S
+S
+S
+M
+freq()
to look specifically for unique values in any variable. For example, for the gender
variable:
+#> 1 M 10,303 51.5% 10,303 51.5%
+#> 2 F 9,697 48.5% 20,000 100.0%
#> Frequency table of `gender` from a data.frame (20,000 x 9)
+#>
#> Class: factor (numeric)
#> Levels: F, M
#> Length: 20,000 (of which NA: 0 = 0.00%)
@@ -403,15 +404,15 @@
#>
#> Item Count Percent Cum. Count Cum. Percent
#> --- ----- ------- -------- ----------- -------------
-#> 1 M 10,390 52.0% 10,390 52.0%
-#> 2 F 9,610 48.1% 20,000 100.0%
M
and F
. From a researcher perspective: there are slightly more men. Nothing we didn’t already know.bacteria
column now consists of text, and we want to add more variables based on microbial IDs later on. So, we will transform this column to valid IDs. The mutate()
function of the dplyr
package makes this really easy:bacteria
column now consists of text, and we want to add more variables based on microbial IDs later on. So, we will transform this column to valid IDs. The mutate()
function of the dplyr
package makes this really easy:as.rsi()
function ensures reliability and reproducibility in these kind of variables. The mutate_at()
will run the as.rsi()
function on defined variables:as.rsi()
function ensures reliability and reproducibility in these kind of variables. The mutate_at()
will run the as.rsi()
function on defined variables:eucast_rules()
function can also apply additional rules, like forcing amox
) and amoxicillin/clavulanic acid (column amcl
) in our data were generated randomly, some rows will undoubtedly contain amox = S and amcl = R, which is technically impossible. The eucast_rules()
fixes this:data <- eucast_rules(data, col_mo = "bacteria")
@@ -435,10 +436,10 @@
#> Kingella kingae (no changes)
#>
#> EUCAST Expert Rules, Intrinsic Resistance and Exceptional Phenotypes (v3.1, 2016)
-#> Table 1: Intrinsic resistance in Enterobacteriaceae (1288 changes)
+#> Table 1: Intrinsic resistance in Enterobacteriaceae (1278 changes)
#> Table 2: Intrinsic resistance in non-fermentative Gram-negative bacteria (no changes)
#> Table 3: Intrinsic resistance in other Gram-negative bacteria (no changes)
-#> Table 4: Intrinsic resistance in Gram-positive bacteria (2832 changes)
+#> Table 4: Intrinsic resistance in Gram-positive bacteria (2750 changes)
#> Table 8: Interpretive rules for B-lactam agents and Gram-positive cocci (no changes)
#> Table 9: Interpretive rules for B-lactam agents and Gram-negative rods (no changes)
#> Table 10: Interpretive rules for B-lactam agents and other Gram-negative bacteria (no changes)
@@ -454,16 +455,16 @@
#> Non-EUCAST: piperacillin/tazobactam = S where piperacillin = S (no changes)
#> Non-EUCAST: trimethoprim/sulfa = S where trimethoprim = S (no changes)
#>
-#> => EUCAST rules affected 7,405 out of 20,000 rows
+#> => EUCAST rules affected 7,442 out of 20,000 rows
#> -> added 0 test results
-#> -> changed 4,120 test results (0 to S; 0 to I; 4,120 to R)
Adding new variables
data <- data %>%
- mutate(gramstain = mo_gramstain(bacteria),
+ mutate(gramstain = mo_gramstain(bacteria),
genus = mo_genus(bacteria),
species = mo_species(bacteria))
AMR
package includes this methodology with the first_isolate()
function. It adopts the episode of a year (can be changed by user) and it starts counting days after every selected isolate. This new variable can easily be added to our data:data <- data %>%
- mutate(first = first_isolate(.))
+ mutate(first = first_isolate(.))
#> NOTE: Using column `bacteria` as input for `col_mo`.
#> NOTE: Using column `date` as input for `col_date`.
#> NOTE: Using column `patient_id` as input for `col_patient_id`.
-#> => Found 5,647 first isolates (28.2% of total)
filter()
function, also from the dplyr
package:filter()
function, also from the dplyr
package:filter_first_isolate()
function:
1
-2010-01-02
-F5
+2010-01-15
+V8
B_ESCHR_COL
-R
S
+R
S
S
TRUE
2
-2010-04-18
-F5
+2010-02-08
+V8
B_ESCHR_COL
S
-I
+S
R
S
FALSE
3
-2010-05-05
-F5
+2010-02-24
+V8
B_ESCHR_COL
R
S
@@ -541,10 +542,10 @@
4
-2010-11-03
-F5
+2010-04-28
+V8
B_ESCHR_COL
-S
+R
S
S
S
@@ -552,32 +553,32 @@
5
-2010-12-16
-F5
+2010-05-13
+V8
B_ESCHR_COL
-S
-S
R
S
+S
+S
FALSE
6
-2011-02-23
-F5
+2010-05-14
+V8
B_ESCHR_COL
S
S
S
S
-TRUE
+FALSE
7
-2011-04-26
-F5
+2010-08-04
+V8
B_ESCHR_COL
-S
+R
S
S
S
@@ -585,8 +586,19 @@
+8
-2011-08-01
-F5
+2010-09-12
+V8
+B_ESCHR_COL
+S
+S
+R
+S
+FALSE
+
+
-9
+2010-10-03
+V8
B_ESCHR_COL
R
I
@@ -594,42 +606,31 @@
S
FALSE
-
9
-2011-09-09
-F5
-B_ESCHR_COL
-S
-S
-S
-S
-FALSE
-
-10
-2011-09-11
-F5
+2011-01-09
+V8
B_ESCHR_COL
-S
R
S
+R
S
FALSE
key_antibiotics()
function adds a vector with 18 key antibiotics: 6 broad spectrum ones, 6 small spectrum for Gram negatives and 6 small spectrum for Gram positives. These can be defined by the user.key_antibiotics()
function adds a vector with 18 key antibiotics: 6 broad spectrum ones, 6 small spectrum for Gram negatives and 6 small spectrum for Gram positives. These can be defined by the user.first_isolate()
function will automatically use it and determine the first weighted isolates. Mind the NOTEs in below output:data <- data %>%
- mutate(keyab = key_antibiotics(.)) %>%
- mutate(first_weighted = first_isolate(.))
+ mutate(keyab = key_antibiotics(.)) %>%
+ mutate(first_weighted = first_isolate(.))
#> NOTE: Using column `bacteria` as input for `col_mo`.
#> NOTE: Using column `bacteria` as input for `col_mo`.
#> NOTE: Using column `date` as input for `col_date`.
#> NOTE: Using column `patient_id` as input for `col_patient_id`.
#> NOTE: Using column `keyab` as input for `col_keyantibiotics`. Use col_keyantibiotics = FALSE to prevent this.
#> [Criterion] Inclusion based on key antibiotics, ignoring I.
-#> => Found 15,789 first weighted isolates (78.9% of total)
-
isolate
@@ -646,11 +647,11 @@
1
-2010-01-02
-F5
+2010-01-15
+V8
B_ESCHR_COL
-R
S
+R
S
S
TRUE
@@ -658,11 +659,11 @@
2
-2010-04-18
-F5
+2010-02-08
+V8
B_ESCHR_COL
S
-I
+S
R
S
FALSE
@@ -670,8 +671,8 @@
3
-2010-05-05
-F5
+2010-02-24
+V8
B_ESCHR_COL
R
S
@@ -682,8 +683,32 @@
+4
-2010-11-03
-F5
+2010-04-28
+V8
+B_ESCHR_COL
+R
+S
+S
+S
+FALSE
+FALSE
+
+
+5
+2010-05-13
+V8
+B_ESCHR_COL
+R
+S
+S
+S
+FALSE
+FALSE
+
+
6
+2010-05-14
+V8
B_ESCHR_COL
S
S
@@ -693,9 +718,21 @@
TRUE
-
+5
-2010-12-16
-F5
+7
+2010-08-04
+V8
+B_ESCHR_COL
+R
+S
+S
+S
+FALSE
+TRUE
+
+
-8
+2010-09-12
+V8
B_ESCHR_COL
S
S
@@ -704,34 +741,10 @@
FALSE
TRUE
-
6
-2011-02-23
-F5
-B_ESCHR_COL
-S
-S
-S
-S
-TRUE
-TRUE
-
-
-7
-2011-04-26
-F5
-B_ESCHR_COL
-S
-S
-S
-S
-FALSE
-FALSE
-
-
-8
-2011-08-01
-F5
+9
+2010-10-03
+V8
B_ESCHR_COL
R
I
@@ -740,45 +753,32 @@
FALSE
TRUE
-
9
-2011-09-09
-F5
-B_ESCHR_COL
-S
-S
-S
-S
-FALSE
-TRUE
-
10
-2011-09-11
-F5
+2011-01-09
+V8
B_ESCHR_COL
-S
R
S
+R
S
FALSE
TRUE
filter_first_isolate()
, there’s a shortcut for this new algorithm too:
-
date
patient_id
hospital
@@ -795,12 +795,41 @@
-
+1
-2013-12-28
-M2
+2012-04-28
+X6
+Hospital B
+B_ESCHR_COL
+R
+S
+R
+S
+F
+Gram negative
+Escherichia
+coli
+TRUE
+
+
+2012-12-23
+Z1
+Hospital A
+B_ESCHR_COL
+I
+S
+R
+S
+F
+Gram negative
+Escherichia
+coli
+TRUE
+
+
2011-05-27
+G10
Hospital B
B_STRPTC_PNE
-R
+S
S
S
R
@@ -811,13 +840,12 @@
TRUE
-
2
-2017-03-17
-Z5
-Hospital B
+2012-08-19
+Q1
+Hospital D
B_ESCHR_COL
-R
S
+R
S
S
F
@@ -827,30 +855,28 @@
TRUE
-
3
-2011-05-13
-G5
-Hospital B
-B_STRPTC_PNE
-S
-S
-S
+2016-05-06
+Y2
+Hospital D
+B_KLBSL_PNE
R
-M
-Gram positive
-Streptococcus
+S
+S
+S
+F
+Gram negative
+Klebsiella
pneumoniae
TRUE
-
-5
-2016-11-01
-C9
-Hospital D
+2010-08-27
+I8
+Hospital A
B_ESCHR_COL
R
S
-R
+S
S
M
Gram negative
@@ -858,38 +884,6 @@
coli
TRUE
-
-6
-2010-12-16
-E2
-Hospital A
-B_STRPTC_PNE
-S
-S
-S
-R
-M
-Gram positive
-Streptococcus
-pneumoniae
-TRUE
-
-
7
-2013-06-03
-G1
-Hospital A
-B_STRPTC_PNE
-R
-S
-S
-R
-M
-Gram positive
-Streptococcus
-pneumoniae
-TRUE
-dplyr
way, which is easier readable:genus
and species
from a data.frame
(15,789 x 13)
-Columns: 2
-Length: 15,789 (of which NA: 0 = 0.00%)
+genus
and species
from a data.frame
(15,783 x 13)
+Length: 15,783 (of which NA: 0 = 0.00%)
Unique: 4
Longest: 24
The functions portion_R
, portion_RI
, portion_I
, portion_IS
and portion_S
can be used to determine the portion of a specific antimicrobial outcome. They can be used on their own:
Or can be used in conjuction with group_by()
and summarise()
, both from the dplyr
package:
Or can be used in conjuction with group_by()
and summarise()
, both from the dplyr
package:
hospital | @@ -976,26 +970,26 @@ Longest: 24||
---|---|---|
Hospital A | -0.4824505 | +0.4697326 |
Hospital B | -0.4689469 | +0.4770958 |
Hospital C | -0.4689858 | +0.4799670 |
Hospital D | -0.4772799 | +0.4785082 |
Of course it would be very convenient to know the number of isolates responsible for the percentages. For that purpose the n_rsi()
can be used, which works exactly like n_distinct()
from the dplyr
package. It counts all isolates available for every group (i.e. values S, I or R):
Of course it would be very convenient to know the number of isolates responsible for the percentages. For that purpose the n_rsi()
can be used, which works exactly like n_distinct()
from the dplyr
package. It counts all isolates available for every group (i.e. values S, I or R):
data_1st %>%
- group_by(hospital) %>%
- summarise(amoxicillin = portion_IR(amox),
+ group_by(hospital) %>%
+ summarise(amoxicillin = portion_IR(amox),
available = n_rsi(amox))
Hospital A | -0.4824505 | -4701 | +0.4697326 | +4675 |
Hospital B | -0.4689469 | -5555 | +0.4770958 | +5523 |
Hospital C | -0.4689858 | -2386 | +0.4799670 | +2421 |
Hospital D | -0.4772799 | -3147 | +0.4785082 | +3164 |
These functions can also be used to get the portion of multiple antibiotics, to calculate co-resistance very easily:
data_1st %>%
- group_by(genus) %>%
- summarise(amoxicillin = portion_S(amcl),
+ group_by(genus) %>%
+ summarise(amoxicillin = portion_S(amcl),
gentamicin = portion_S(gent),
"amox + gent" = portion_S(amcl, gent))
Escherichia | -0.7303198 | -0.9056119 | -0.9745730 | +0.7345687 | +0.9051757 | +0.9773802 |
Klebsiella | -0.7343027 | -0.8955717 | -0.9643093 | +0.7307190 | +0.9071895 | +0.9771242 |
Staphylococcus | -0.7304946 | -0.9196838 | -0.9821520 | +0.7353606 | +0.9210857 | +0.9816537 |
Streptococcus | -0.7308142 | +0.7268273 | 0.0000000 | -0.7308142 | +0.7268273 |
To make a transition to the next part, let’s see how this difference could be plotted:
data_1st %>%
- group_by(genus) %>%
- summarise("1. Amoxicillin" = portion_S(amcl),
+ group_by(genus) %>%
+ summarise("1. Amoxicillin" = portion_S(amcl),
"2. Gentamicin" = portion_S(gent),
"3. Amox + gent" = portion_S(amcl, gent)) %>%
- tidyr::gather("Antibiotic", "S", -genus) %>%
- ggplot(aes(x = genus,
+ tidyr::gather("Antibiotic", "S", -genus) %>%
+ ggplot(aes(x = genus,
y = S,
fill = Antibiotic)) +
- geom_col(position = "dodge2")
To show results in plots, most R users would nowadays use the ggplot2
package. This package lets you create plots in layers. You can read more about it on their website. A quick example would look like these syntaxes:
ggplot(data = a_data_set,
- mapping = aes(x = year,
+ggplot(data = a_data_set,
+ mapping = aes(x = year,
y = value)) +
- geom_col() +
- labs(title = "A title",
+ geom_col() +
+ labs(title = "A title",
subtitle = "A subtitle",
x = "My X axis",
y = "My Y axis")
-ggplot(a_data_set,
- aes(year, value) +
- geom_bar()
+ggplot(a_data_set,
+ aes(year, value) +
+ geom_bar()
The AMR
package contains functions to extend this ggplot2
package, for example geom_rsi()
. It automatically transforms data with count_df()
or portion_df()
and show results in stacked bars. Its simplest and shortest example:
ggplot(data_1st) +
+
Omit the translate_ab = FALSE
to have the antibiotic codes (amox, amcl, cipr, gent) translated to official WHO names (amoxicillin, amoxicillin and betalactamase inhibitor, ciprofloxacin, gentamicin).
If we group on e.g. the genus
column and add some additional functions from our package, we can create this:
# group the data on `genus`
-ggplot(data_1st %>% group_by(genus)) +
+ggplot(data_1st %>% group_by(genus)) +
# create bars with genus on x axis
# it looks for variables with class `rsi`,
# of which we have 4 (earlier created with `as.rsi`)
@@ -1114,22 +1108,22 @@ Longest: 24
# show percentages on y axis
scale_y_percent(breaks = 0:4 * 25) +
# turn 90 degrees, make it bars instead of columns
- coord_flip() +
+ coord_flip() +
# add labels
- labs(title = "Resistance per genus and antibiotic",
+ labs(title = "Resistance per genus and antibiotic",
subtitle = "(this is fake data)") +
# and print genus in italic to follow our convention
# (is now y axis because we turned the plot)
- theme(axis.text.y = element_text(face = "italic"))
+ theme(axis.text.y = element_text(face = "italic"))
To simplify this, we also created the ggplot_rsi()
function, which combines almost all above functions:
data_1st %>%
- group_by(genus) %>%
+ group_by(genus) %>%
ggplot_rsi(x = "genus",
facet = "Antibiotic",
breaks = 0:4 * 25,
datalabels = FALSE) +
- coord_flip()
We can transform the data and apply the test in only a couple of lines:
septic_patients %>%
- filter(hospital_id %in% c("A", "D")) %>% # filter on only hospitals A and D
- select(hospital_id, fosf) %>% # select the hospitals and fosfomycin
- group_by(hospital_id) %>% # group on the hospitals
+ filter(hospital_id %in% c("A", "D")) %>% # filter on only hospitals A and D
+ select(hospital_id, fosf) %>% # select the hospitals and fosfomycin
+ group_by(hospital_id) %>% # group on the hospitals
count_df(combine_IR = TRUE) %>% # count all isolates per group (hospital_id)
- tidyr::spread(hospital_id, Value) %>% # transform output so A and D are columns
- select(A, D) %>% # and select these only
+ tidyr::spread(hospital_id, Value) %>% # transform output so A and D are columns
+ select(A, D) %>% # and select these only
as.matrix() %>% # transform to good old matrix for fisher.test()
fisher.test() # do Fisher's Exact Test
#>
diff --git a/docs/articles/AMR_files/figure-html/plot 1-1.png b/docs/articles/AMR_files/figure-html/plot 1-1.png
index eef839e20f0193eb97acf45da65fe1f14305bcd2..f78b6151ecbb262ff2016961ab3634a8af5de5dd 100644
GIT binary patch
literal 33823
zcmeFa2Uu0vmM)5ADHwpQ7NCews0b2GNCpKJ0ZEc`VM0MNNDieaf(nWv29hEfK|~~j
zB8Y&Bft)jva}N9c6VK`M&b!_HPN#Rj+pqb$8{1-U)?RDQG5$2BerhW6^lMqy($LV*
zD;z&^l7?ngBn{2VH*~+?Z%%Y8Q1NYz&2c?@8k%jskC_%NG|Ag}%zwNCFJw4A3_e0_R-b>Ruwpy}xnw?~g3ANp`4Xj?>2N?3Z0@w<7e
z!})q{gSwLJjJO&a_4}1yR+2x&39O*SH;T(I4EXl@*spB(cBpfWD86}r*kprmG3;A;AP$mHOWRNPo9)yuP86K_>p?Z
zl8c)=!?NL^kD9J-rpNr$HFkzd?X+{B_U+pzOP6zLus+R9RgRI?oNxdB{Zdj=KDQ$x
zOf)qszr1^*`10jTR#&=1!Mbn%trx}Jip&4^eww(C_XP~JlXOQ*xg*@iO1b-MQd%;s
zO){*ShH{4U`(CHHPWG8acpMZFahe`#($mvhxqAKPy{C>GJ9cc4kKDIghocxoo88Ah
zm9V=d-M{rQOv0)D)rC>#+_UBP>+0$rAGCb5-|TbY)pfBa9tH&mYdtQ*iw^Dh8n-e4
zzj{P=LclA_|=^stOw
z=ZDVDPKqqXaK>3ilxq0tlNp|bnw}%F~!mluJK8uZweRiOW%0wIIr97)z
zwN2uZ4`tu}{e0nyxl`$xna;z_c>?;c;;gz}O3x3ZxeaB-wrjOem!|U!59*&veUNf9
zW+xX{v4FBh(0%e%{Uyp<{Clp;_%-%l4rzS9$Dv+_M^9&3@YT7O%E|QGZ(VWBl;11oS$utZl%M%V
zNk@*;u+{RG;Ns%)4hxey_wj*$my3%FUKl_7
zBVPW~V-eMh@7nmH)*M$+5u_Jw{B~f)r$@p{zp~9w4caSx|MeMe=q~}6FJE3qquVb?UDcz!!U9ihg=RV|*~5
z&NYhb)PP>v@)C|BS2-~;aaTorx6eo!U&=j!mX~MCgSeDr>Ey1hkJm#*q={6&Aiknx
zpf0WC)>fyUFsJ(hY+`n2PM-YcHk>D7QJvJ@-cBj}xwZ{;b!(&K13ee#s{XPti#h?=sBcwUybbul-y_bIW*ZQM^ya{l{rmRmXISN3_#T5_l$4ZIdY9YG%q*loRX><7BqXHffs8?!`9E00Hg5fM
zxIVpnSJZ(O4#0?L*9s(`
z{z6~cZuV$%#sM_8;&E%;(qQdD3`OmEIgaFjMn8?r`wPbplWRi`x9Z
zdUgK*un>xVLi^=RbaXWC&w0ad-@k8lrSm}dQo_}+v}
zfyoIaB_$OVtJ=3T2loCFAjh?1#}4Kj!V^5_kD$8$ORrTvbc0y{kby?|mUo9*X+?$g
z56&-{7n`yjy^1aS-PjxlX~v)SqW%?SM#poYw4O*BZ%REaudc2x$LPy@q2b_|*Kcn!
zeDGq_U%!$Lda1RncTgv;u@`cF)$A!gEL@e6AOHEr74rlW4~;M#hu|fusjY3QHEpC-
z-I?ET`~M-#^dG&M|8|u2Kfm$6V`(#1lw)jd)lM;997wC^?(QCH&XeHdvv}jZV#l$Y
zA3l5#{c*JgFtj%HG+(?8kZg-<|MO@Wl?ds@_}K7z>rY#)GTV~UJ|_%)0KoR1`&zp)
zAj`f-%r@_>7Bn`y?dXzQ-urcYoB&v}psx1R&=vIJkRv{LvfC^HS+<(?gU(b@^>t
zqom8}>+5F~R#sNB(sN)R#4+9yZ>@b{`feqoBqgvUC9~bX?VT?-H@9`&%d=yhrRaYe
zF()4GJFD0gl3`I})&21yQ+bGxiD!+G6VOrErTPm@6%1Ps4wT3&**2ZPJJQ6l+;U|M
zc(~u}{=It>!!2H&AMTWvmU8-lLp0%oUAS-oZ;JOw(Muce{GzlQL=_V&RIYP!h3%z)i}>JSfUCbZq+*G&hGB&$hwRcPq&tQ57PYaJ$uT;r`>2*E&v<`
z$oZo=Z~n^@x%uf~lPdvuFK^fH&MO0i@@KaDe2GzheD2dDyjQ$NETu5S#8dC&$)J0h
zc}IDyb4QQSng00lWcTj*i20rfuW-@YY{&lhx3;qI>+bpEjkn(>ZtX@FQfbcbWVL>i
zdb%WlB75%Kxnd1Jry^z+mf@zH=|jT`btzyB3mqI(I$CGxXgOa7z$od_qD>N)25NzB
zZ*3JJ2#TO0fHQXYk+%_x17`D<_KS0qUQ6QCGW5GI^_rN
z;x>cw`vL&*te5uCD4*z{7X?RPDO=d4rrmqWAK23TDxF`~)+XFA(1PMLaHdOj9D$N!
zA4WS$oCa$5a&wzC=en?%xZ8qf@EQk4L`0l_9ok@(BOoP}*LOOq)FNS8?w@H9%RE4i
zaYbbOh1|zCcP)Jp;^j?nP5=$DReN5LcH_nk0Q>@9NqA%^Q@PR}A?coRkX4?v|!j+x@k@=)XG7
zf205q*cZ7VtR&H?+S7dZd%C&?oo3gx=qpcpo=+HWk9^dp)AF~L#Fq_Z+#Z>kS(~v^
zErWG<2QRuFdMH>&qV{uMC-bhdU{0TKDL3uB^o4~5Fwv{#(h(bFua|UK)Fjh$I$+J|
zvk*07GG=r&TW@bKcbLOTy0%~)$jYotZv5IKvTrNo9wT#E?`pp
zK4I7F+;I!6!@ZHHPQV4Uj=&f*vJ+#rL{Vbr5_oa9U}kfFLaytcf0
z^{QpVYpK4tD6+6LRK&QLZ&bZKgLn6qi%O+s!eP2j98#^^2V7&X4^D
ziSeFCx+e!KePkkH#M5KMQ?gG|BUBa3s0+hR!190Ih
zI3sf@C&?n(YgA-!x7L2E-ecvt=5H1o9|%{)XGQbrP+Qziexve7szCdYis^gQ=aq0M
zY=pMx$xFZiXGvq~;<;+C5Jv$B9Z!61jN74~ue5inTuJ2HLlgJV!3OJvM^z{Ge-@2;;=7W5nLkjZQu3zimm!Ad
z_$Lt=)Lrr6Q=db{{AkYN|2j6$JYnvAtFfO@Pkbc|-jQct!+GY+8KOV22a?KGur!&{
z@3OvS_Olb|36w(Hl5I7$`W6?P)ESpm<_&j4LZ%DX3Heao-`tXRs@NaEv8%JQ(WNI$
zL18lccmXP6Y`Cj=$HnKVj`C8=4E9jbi3sVu$jIM)Y$mI8Ogtv5IDIOl=Pv>(&`a{4
zS0iN8`?|}-S9v))6io5nXLpYUu+=?1dSfKTS9>b4(L%ZVbC}!GY>!i4oyyw`nZ5^WmQ6zA9wi7_lTtlNc4QWcZ+>-zi?>zBk<_GzF0@g;ZBj=y#L$@I?nz3
z$5G)KIHA&-L9Y!q*&0}4o!NHH{Plcq$K74Z;&{~D%8w=>e|s)Y)dWE~Y79*(R
zUaJ||7eWdax`w@|^GZT_2do-1b@^WdkmDMx>R;_vw*-g5?}^@0w7)`9L-Rd^;{?%4
zK*Pc45Mx$xhadRsG`p@cAn`^pocQ>7NS|AT%_}8_b2WmG;a&*ljuZk~?)4cNrY;Rr
zCrzm7XG((r*yodThRQyEQI2b<~)`AqN!)^mnBR_pn4csf5FdG%dIqW$`ggBNQ@
zKI{lI#JUJJl(ufpU9)M|!f2q^&EivjLG~3=8GtI}G60r2JSM*4Gd*XJ&%rO^OUWHe
zEn(*ywEA$H(=5!fI;$t#KvUCUFuevstZ#HIL+2A;c2}s&8*4s?Nn{)_5$(FCfweDo
z=EH3$=#cws-uW`h04`d{bmRVa;j^vdbQo@4#M+Yp@K98xXP0yys7(b$0a#1%kgpV5
zZsyxNuyQ|DR4{VxoMKjn^bWefz{zitFF86<_U8mZ$#enLBkM$4_hV}-5I8!{R5|%=
zWo2cL4qEDG4Pk|b+IGCZO3_X>oW+l@zq&)jT%X&I9*LtW*(GvABc9D^Srqh9Gc72!KG<~Q@Tt^V4OJ(20&un8EusdrUV`3VUNCmcoYJW2fj(Ml4sWuOs-YUH2t#vI_H}ZVbc^
zHRVLumF?#A0i^tmnVCKADj?cmMlRY2rw`a=O_z<_X0S5Pay-q)_wO^y#<}My(6zl|
zlsUn+9=E!H>D#w&8x@K`;Vum|>W!yfqsZ#!ySucvVzDHi)u@zN{=f$$vBO54B^L@R
zV(%$l6n+#EA_Db#W%_OB-=d!uNyz)(xY2(mG+O-9KxWGr>N%kIRSG;eUh5~X3Qc{t
zU5XTscDuv&D*ejbTz|?YTAz)@ItC7I^%0NsoxtloftWawxkd
zOJgA`r7pY#{iaS9Ea|}bP*M53&8bnnoZX_p-?TJ{%daNq-uTk|F!#Z=3Ky;P{WhtH
za@vl3mKYw$kTo@JgPR~r2S!=(;RBCAN?ofe%74*Y1M9rWa+9T(s|CJzFF(m!O?cXy
zBrMjmiH^w2%zgbc0W_}Bi@IpyK6(_taxkOCxZLNd%Xqin6}Si0Nmof1f-J#g@@x!+
zjho@Kcz;rv!H@d%er(Ac;BwN5(ki6;Q9HBNs)CAuB+AnDgxaeN(ni37aUN+czQIxf
zQ6Kxm_}NAQpmC7&0a!EYYHBF*ph8?#@Vf*d_EQS+Ha!&)CgmzQ4!uF5x@qT>Kk3h4
z>!2cK+cj7&3|f1+{dnRlyWHXkC93~`3dJVN!_UtqbOf=p<`f0_>hjEcwb*ZhYFGxdM0;^lq_oSCRm38j3jI{JUOSwWM
zaVxbhn0|Yvpb@L$(RP)EnO|4!b*^68x$m*690$&;X|ehhpQ@0WZtMy%SwcUCs%Gpq
zn0~eb4%1?vPC-Iu+i}a)RH#y^%8w5g%zm;aG&ojfu@_K5L+ic5l{6yxN2>ERr9R~q
z^16kpZ{Bt$G~=&X0H`71V)o|*a`M}*Z4LKYnng>P#*$N2R(?Lzg+>Os4ee756j1^j
zDLmY>^Nu{Z1U{YQ+|D4Sgl3gv$KFHJApY7-HnEqW?_ld&wrs(j%^k|_mqxR~B1*KP
z<+xEilso=8P-ba{rC{bLgeuU`$DyH(30jFji#IuBmawKI(HIh%StoaC7l$xv7@;-L
zkJX)98uF^~igmOoDWayFGWDIVNTlAWB))@g>WEBuMFRQ%KnU%RH#jMY=v(!&o==#z
zw^0%~Wz}1@?Z#R7>!IE`>_D}v9>Q=y0mDh+cC8L
z2vG~(Jo$!}l!hYr?5@h(kW&Gw&^Ka^TQzufKQsf``3j8^I#G>@mq*9@Tcn>f2r}Gh
z0S2CQF};;@gR(3n*u-HGc@8yZ0VQUS6t2UHRqGk->Fo5OkV2O4F%PYlsAZ1LtJFDL
zalO9>5qtz-dX;hv5|Hb}kCBwzsHg83B=Fgu8)n&IgR>YpsQWT(!wT|Uzhw*!jhh!S
zgCSr^h__0(E4Y1&XkwZd&-8o_!>{gZ$gqY63VKP_x(W6vR$VmHt`iR^H;Pfa)6FWv
zU$O`S&LCj|`TA9kjkl3Z$LI|1meZZpqW9;3~|Et+X-^pgE(SkwnJn_c!r{AO?AMq-8-jSDUJb;`7vt
zy5t0+0jgkAv_YN?m@!c&W$BZ(=l2wo0@P2gzJ(sIr5+&cO!KOQ$JY5-6NONz2vvdR
zlmST^OhK>I;nHxkN3%=Mt<8JeQQR0fCwjkg3kX;uoIrj?p2I^kQMK<4vygFPfftp<
z*!c7Fyxf
zL+-=vN;WC*EV{nIkBbo(gT{LXy1){CPsPLi+_r7>B8@MgrK9C+bD-XVuv`G51pl2#
zWmv+^z@e%#iwA^+tothC@%QNqNerDakFB!1A6eut&Gq9d!9knwW1K#?vgdkf-IPh^
zAIlr^dN|Sd2kX;Ol?nX6w=*aJog$ciS6N?60hKg0$cerbxS7<6SefXUm}I0Pz+Abx
zV{z{|eUP#MR_1S}-3ZWicd_Ue0ONP7(E(_Opy)3)4+fzW@J~LwZqNVd#H)T&wVzYu3*L0^P
z96`}QPk#hsV#XGI_6jXyKuJgRet!zVn>1hV>c;N{`TJHy&thnE7=a{!nW;wmioxk1
zEYjBlfru&L6VP+s+_HZcpzte6Vd2T5O{(iTiKlS<-p-|TulftITxCkz94pnydtA8c
z6S2_AO}4JFaq~#qI}qoHYZQ{0`lj1?C>(2qQaGJIop%?HFw#Cn4%)6!R!|X5b+VB`
z#Jjjn%O3Lfb4Xr56K3Evbga^fzKSOvV(fZ>uI;NlN7JwtQ_W$(6`hm$_C24A07Kw`U}_*GMve~l&)Ibxu0Ng#8fE9s59OfS<)CB2JaPK6h3it?X6(Vk7S+#B2Z+TO
zmI5|2mwl9yT?ULtJ`&!Kw>@ZCAL`WXa<&}&d6Vn|XuJgZ`sp!t`)$)y+IbEQi
zZq6jjGyhF5L5Y;`L-*Y0dSg*p7&&oIoRT5|M>u_mBI8(l`V{RgM02zygLzQl%r2Bd
z?zYPAjiD6cc_OU#!m25|AE-XkXlJ;WUaqsP(XCkngn-DqA{SQFF^V6stdD>y4!up-
zqWWI`%IGQ2kt-ZN6e9F6O3y@IoJ9xOm~sVB$$k?T3QY(K4xkwBwupIU9Cx_$b|@O7
zB^+Kb9{Gmbfu#b9PpxC-cSh!?&%Czf;PMk^y^6NegJLcT(#99RDX?TyV8l&!sq6`(
zgtI7agKjPJ#xpgi`5jMA5@G_t{M6>%HF_AGJc5EtpPlmeV?iHj6Y`oZ
z$7N6d_}cMT_UnP2&*_8G;Ds9&>_DpL*C_dtj!|L^4M^x=?c=Gx`Zk@Vwq|*I_?FF0
z_NOg%q1MxJCK+bEeASszuxMFV2KS;4oDv_A?}E+J!Yq=3jGVaGxkHZ0Q2Qx`V6&vx
zh(A0^*M^TiT$k1ZtOfGxvM^dQ-TS>KE-FADZvrW87Ew2@*A31PRH!p0GBMH7Qt;#tgm*)NMKwuF;e?lGt&sw#>bFs{L9`%{swrB&2S~hkXUoUBC
zB*ugV_8CANJq~UtI5WeReVup^4Z~Z_cXR&)UWe{FqD<*5UX3b%K?en>MaLM%a8G1K
z_)m2)?V>`v5jT^7VdjgzJp2`!-!?x-9;UNr&w7j&v&>6XVd?o-#11%>e?y5a7a}!Z
zsKr`>Hv7dnF50;hX#9Ij3w;zs2j_f}n~g!r(JA3x54%!SC;ij}Jg|7}lnowk^p
z=v9Rs(i+0l`QXvmqf6s37{~FvZJ_ef*$gzwmCMJgc=@4G%?OKf(=RXqH
z)JzQn4_?HxPtN%3D@C@1fg6ApQbei@h*>}%I*|uVz{UA#8HXm60-DJr7=eD%{m5;`pkQ^WI6_bL83A?x
zsN2c4&3!}JKW|XqZ*qCXkVAq`eR}XZfvz^le)#w?$(eWbTQ!VBz|`00zdj8*2BlyR
zR@x?2N#bKek!s1b?O1lD6H^3XxW1+3Sg5P;)Se$Aq-HUH`wRdCuIAI>SiPRL{RQRO
z1W%E0j;XP5$9=0;9}`zMR@~Cl2zz9ZZ8u5HL-i^;3^BDot3T#R{8H@NO22GIQEBt?ExQ2=yKH3umQKzY5ihur+IM|
zN5)61DsNWM#~qpcB>*>qh%x|@sF4eR>*Zm$N;=TP7YX#!-v=TMOg6DFUg6aj9}A3;
zSfzvfE8Ua0xVZNC!b0lo)x(YrCo%!MO`O}VVplIrdr_S(yixBMf+S3MaB41i40~YB
zAI=A=Dv*8hUoiQmJUO2rr#!jCPRUC`xUHf@U~y0rmP>Cz#nvB*wn{ZVnya2ZbzM-wd)>g923UYQ#rD5=FPhfnABze)QITB``V5u`hSBi2
zWnq`CJ4ihVSfVmLI&k46IDs4;bd3>6+qi+a&}B-BLLlJ8o65OtyLSl4Y+}Khu2glx
zntj8mmB++}M@!Ec>eBs?sb=5DN5Xum-%O#K;*D&c+Mp3ZPe(P@g|LRjdq;Os{V0$e
z>~=nWenZ@NQX|9lfzs%npdD1Py2m?9=A#@_WWj6to=Bih=p**Kd+8GgSZ67F)swG@
z%|Slsv_2HD0h7^6h`}3Wk7GrJoC<{lK;$Vp)6w?gsy;p`sYS3hUdyaXq-5{&>ur=J
z<-JRQ?Z=utA_JjEaQeJFd~A?xLD3^1w|%+EODV)+G;Df%i%o3b?%loMLXU+lwBe!Cnc_~~3>vGXLz+Sbk4LoF}?#q1N-&Nd&K8ZD}XFL$mX{0!>6SmlyG7A2obk8ATOOg1Q2I
z#?Z=Y2v8sNb`Jz6;#lk-Z=Jv7G}Jf?IeCf1^GTz}FZw$%jKP}Kw~7r$zad&K$emiUdt2dEzblJY6{)y06xmz*FaJLi
z0WZ0r^T5?W7c_C_$tES1HL!2+?I?6?Zl4kS#$clW_>>dC@>eO4i%EPMxAsyCluV1~
z$4EfbhNfzr=*C3OrP&HEW4LR6A%(~dwIH|@;kD#KX9|QcpU0a@jXS$xc{iNQlh+i?
z>I@>}tSUhZq{F>8IwU~6FAn!-=OspE1%>!bOEWrCL^J$Yio)^W9rvBWOTFz9GA{2B
z7Is9o3vif`6UN~nTm2n)S?af3+T1e
zfZ^W(p^hjZLj);uiGw=I!NkY%+i#7_l#N_eK?A04@Q{rs+0_+hGQ8$9LYKd=nvg}u
zW8;WB=ju8~VpccW$ZudXRQ@I4Bmvrqw#%>7P3ItDEzS9h%Vgm1XJZa^P7XCmFyA0>
zq*?Id&mUpq#li@abQ#-X777L&_UKVP))LYn(wNdfH%PoYRg+@kH9wfa$a)|S%?;_a
z=lbd5__RdV1SrQxs8nE~b;M_E!OkI8T7Rqr!!vDbaw*;`XMwN)PVR~VH50^!Hr`-e
zusGFVO`KOS6|6SPTU&)-W8p8UBfc__v#wHrM(Q)H3((cTZ1Dx>0^klaI6d<%RL`ap
z-EW2ps7vtw!ckR>vK_=KJ+#{YrV-^C)z&VL+q~gSg&l{R7r0eq=$1%xEnb3D%4@48
zaZH=wzD6sB1aQdYAG~D(oq;+VLDe%fl*EFAdSVTz3B$DuWq?fnz{!Bd=>!=G=szwo
z>wss$;-EEon8cZzykvU%^l~cQ_Xu8xyVe3Q@GS_C{6wheUNN!R9xyP%&S1--gjPj%
zK_m8>8mI$If*`|+VVF7%+}rHd%j*O)TQ6T-$AS`){M!C^@#RQM>uti#>KvjfiFGIo
zwvenxlZqPz;=%1>1NkqMAO}sTV+l_HdiXu<^3qaB$jhNB{#CL!_up(S)-dAS9^i*IF#C!T};uFw1NSK+6Mca1n)-5J7^huK?{{hT7osEwunPCV1!i2%e&K!)8UnXsbmF^e1>A4kO&%RDv>a&j_O`a8
z;4FG-YA+zum@Krjfciv;JDNW&1T!b&CnVfg60l8X<}D-n7(6Jy9h+!3ajOt($l%ra
zuV^vwa+ZsPz;R-wO<+Tyzx@Bg67t
zwQ7})kz*C1e$@s4qIR@P-4x!fn3$N~pcwHVx)#b|;y#YEDQ42)*MD^$@Pjl)sF*NN
z#<8F$2kPvxl>Hc=ghrqQh&0-~!|lEct`ZU9Rd+T7`g9X2Gt?*X@sImOEbH&_M2Om+
zqH81WIZ~B;4~KzD$z}tuAxTOU;bC-b5M^Q_07rj8`d;vs3*i5rd_||4qg!W3
zH4>cQ;rUQnsxY7fH|fBIZ_kJlEFkyfKJ&}hK%M(4UO{quTZySWMpg?KJKr;
zCJjZr81&XKYn@Qez)OO^VTADOcs4oK1oMeOcVrPV4y|>F@&|LmGb8Fa*a=EDq!F+J
zN+C*+N$MDEb%Uvrfq?L;7I|<3l04-(`S+G!2*RFsa*FM;>O9>
zsRMs9p}pz>l^HjbbCkjbl%vD8I1Ehmvt9SHw;gq`?q@)Oja8ZsmH}avRF6M#&qE^)mUCPA$s8Oc9>;Pigw0#H9(mm
z`ZAg=(l)U~wsV{K@fGY94;(9p-Qmk&?BcUJbv4XUW+DRjnBEoZL@K
zIJ1@o%8XgKOE(sUJLGocNoi@fIehtZ;vnksa@4x+W)$0KN`FSPtHni*gVxRNzen*}
zKx73s6vfgZWbEw~+KlsGw}aaabkyGRxgmeN*zAWp$Fiy%Cp+qJ8pDu8;RC_@O(vuL
zP$-JRsLJwINvT3-q!dE8W3=f%JJ@gdc&6%DyX^@Df)F-IUwhd;FM%mJ^IC0zWkV%0
z0a0`!uYxP8{J?8?PNmo2IzRYUGu0oKVC!!qrSkl){b|9V8
zoR_2N=C1z*pfmU)KR^FX6YKPZp9J6n9}ilw9k+sgIF$oDl2mZ7YV`qF8%4abGx<>+gOp9>5nmKKnJ~ro|re~f+GY}VDy1^
z2@{RaA(Nl&dCiY_)`j*WKjMaPpc$g2!PtmUR#LJ6J#go}HM0-~(B*+Fj#4sF!$Al;
zPy)8X974lkkec`cYydL?3g|Toug4$7aq3Yd7Y(H9g!csoR>k-&ZipuUG*rASl%Pfk
zW-4$xU>hBe^IRn9Z
zPd{&1-rj!gQ9Fs|ch;IjLz-LwIS8&6f|7!junq1ux&wO1D}WP7hek*(gALl9JJ$_O
z$Jj0y?i0ofRKg`^0SGN>U$g--XF;ezDKG}eBOYI0LL#q7(48IfgzsSPSr-$;MDgzZ
z9t(L%+~;;s&=|DRLCBFq-=fPQFd-5ohWr3J|3Q#
znt~;S|7Ebjh^r)xXwh9wi-`#TG^5SB#-r22DXEh3(
z$`3Ae4{BG*1Qogd2>rD}*vLO9IsKjkS7ntegJL$b#<#8p9R;1RUCNJ&DoUZw&l{&u
z^rG)MbiFdy;v32UL3q{6UD--=Y4HbE#@6aqnpO$1|D=h%RK0
z5<(L{c8e9%XdZ4ghF4Ei?n(x;S^f;qu#9#ksBY{+rQtwX1fAQ7Z?$5XtBCAwC(LNn{&{)wqH&
z>k?^qrvs&km+*-F_^&^Uy|yo#-EkOfvgE7-@!OtIv^rWAP_&k<@8R62BuGp^8+{PD%IRmRgD
z5)9@kem4AX1yVDuSF~QG{cIprWx;$%us%Q{xe|k8$7SHwm>>m6>^&$IBykDJL+L*M
zLxfrK@59BMgOoUq@Zn_Q9?nCaxBs`463fIjvFu(Yv;o
z>9K}ahA`*gp(!UXaPI-gIdg=o;?ENxKCn6n1SP>RJcF0(db+y;B62Nu8
zOiSa23xSC(c*r>eV}z-S+O$Gd{SK!Wl@?(!JnjO5XK%2OVKF8h!>@SkltO$y+v0*r
zJVyU1on=$b5sA#LB3ABLy%nMzHz|cFo|B1?x!8~UQpaVen`EzZD;!5M$Jlue*DxV1#J;rQ@b
zyYOs?5#)kw!unw9qpO9ZdEsp_Bu^a+DY$1eb8V}YGd>eRlV~8;jhVIxd85hYz;+rR
z8xw@D^YZRva3Xrb7YO^gbLUQQ@FeuMYZM|hL4C)NKl7wAIX(>PzOn?WExCa
zw!#u7v7xaMex5=?JZ^fa>qWrx7%v(xC*Tvz;pRx_~8
zAeL2Pwo*Y!X&$b(E<{~C`L3?6ThKkg-If(^vIDW!aEBw{+_)D*J1DNW>H^XW@VUWy
z00BZgRICYP(j{ERI-NVfH%CF-IeqqkHWNh$wP+i*4LT6vaoBAu>sl5k$(6zS2bKtj
zj4g@CEgNzK5RViOA3hv^6_Lz-IDvu?gJBMdgKc-*LW=?ME{Y4+}HubG}b0B>4bh
z#d>-lqiaK93aZUQ4GzftxCbP@!lC2vc2$eXXnaSB?A
z+N1Z02#qp|2nDqKZet{DARmskpxQYMHH!HVhKqHIT(zRzqvUT_i!A;Q6!s2M`90r0Y?))k%$qc5J0{Tr6*_*c9c*$
z3zEGiUU@*25Wc}ibmAWqMUo<0*ZT(U4fix(4dZ?Z83bRc?=PfSq>5vice=1
zu`;|Q+4Tx0*h~S0Do!7?u>0E-77RSx|0J5?0G^fPZ$k`^u@8a?`gi?>nZR6`kJ6&rcFv7208AT&wiG8KDUgSyv?-$0cRd=
zBQkQA>-C5TGHwHy5L6<(e#^d{K5CkpzeRW*P;d8p1F%)r{0gBRwO#w1F{#mlK=usF
z66vX$3+e&1?9PA8Q(GV|$kIV|(TbnKwhC_1lJu_ak10R+o+TfA;q?3qtjD!Ovh+WJ
zMH4`;EFFRf!MaY+sgZeOP#T@c1snXNW?!CMq2GA?{7O5q}Us2-S(%WM$Wz;uH%BfP#}>Nh@y;xgwB1alzAbi7UjA
zgg5t|es2gtpCEA{Y-9Wp(S0uUEAf#4I8r_%fXaAr{F!;o?Ut=(Lwdp(9hSf$(6F7C
zHjnPNfTI#tYc;L7+$@4gkA#&o4yo9>;k7_B%s`T;=0J!HdW-p&$iwk`wj>MzD^jGwY<>FpaRE1@yw@JT32kgW$!xfqlq0VsvxYNw
zs@`G^a>SsRR4;+faDAsu!vjqauP45vgCSp7GM?O*q=z%v=
zNL~X?Rg{mf|7E#gV&iTx+I`(v9?F0xvsD?^Xa0p%&oNgn&Vt
z*T3ooXgCNJa50%F;T>d}OVlV*@Nu^Ui$B
zu@R}Ng0Dqo`Lh{kw;FiaYS+C#UspNY^zv4LOODoL;LcQb@mTX%4ef8~_k)ssvh^zt
zE7Y7)%yysW?M?WO4h=*`DTE5OZtdCxmwv@;J0$s-rVX%7g^W);f4f4`RzYQybIlPl
z-bDs~a+y`05gEroXU6?;?y<^gq~pnuKS8eW3_kgb4sd1`~+DY@~=RTpj$lf4*-V
zniLT$Q1l->e3%7o0@ZeDaXyeujHG4>o>C4J99?eoYP9L*RegWbvd8tRYjW5V7WGeo
z=zsM%5yM@~glb~U-?cp`l$Gs}-Aj{__`&5jO*euM4nS=3J<3~OnU3gZN5XfACMe=I
zK+aF_%2`!aPDJ_rR*f5a{uTwCHP95-aUQ(sHQSTVc$hMp>0h%J|Xt-(U^7TN6m_o$_jvZ
zsW2$`>_5WED%6U4P0NP!SX>~Uf*wH4hqjxBcO-n`FRl>4wANBKVSxyEw|l`Wz-dl{
zXpTrN01}eKHW{YOi`faxt)#3hP(>J%&j8spYE0&iZ|*#NNJ5V|*AFd?^dImQbY~Ap
z(KwV1jh0L|#w}D~$`!jA5TXd~{?PRu0gz#ufU5w55$qd5%vlB$pSY5B>DL%JJgc;H
zv_3Hq!A$F^cE#IK5I$xmCVGs1kpbTb^>1i|)VUO@qO$U*4<8WeHJ(j`FFge%8#ke!
z2mnQlW#=p7z9NK0vV;_JQes+KTB4~C-qbrm+ihh4Y7m+uzt0|w3kMAnM6q#J2(#J==hJ1w?vzdFx-8$`xK)hgVmI*1P$ap_xKfxEmCN$pxdn7?gHcML5zb
z%1(-)a}I!S9=HAigm;lyxqze%aK%9N^M*Ctp(JopW=V&)iepa~X@%fwN%sW5j2KN;
zeK=u;j`+lop-M~VJt$?6lV$0!HvFeG%HZR*dtF3|6v2Toz)W1R6}a2W{eka+-7z(J
z-_jb2aGq=W^w9H9eSM4P3i31h`i#yLQ~8IAxIdi76jj~hb{(1B{QMt3eX@jL?CD6BA=JNX6k^4>USyFyQ??=TcP8a-5Sev_H{6ht%5}te(t6+5Dc4i#
z+x{rvl~<_IpGw?VX>X~mVb7iPp*=GjQK?6i$Wb
zDmI8fU_p(j{s0j+i2sIUJK_uvE~-@LZMo#E?nhumNHHt{THz2vk~hrtn#ZlC(Y#T=
z)rO$8pIHC=*ieS~i@-%*kJZ|pnv)&T&nyJc!}_ugYM`M#<9mV|
zIC@tD%LU%$Pbg9rzK^j|1)uMFgzw85zAVh!-~dR(x<+drAx{T`0H>soFrrh7hf{Ij
zJo=aPz>9YfSkqFJk{$XgdtWzL@ih?5Vn#GkZ<}as>R&Fxe_{UYKhsSiCvIdEg;qE<
zr)128>*8$nj*H6+@f*ZZQgo)sx12p)KeIqW?kTP7qcGe6kXeBVNpheOaXY*!oNro3
z@7jI|w=z=J-1^N|>|ZDw@JFE7o)m%F`Mb{-NJ9P-8lhGW@c2o1p~X)IL!!&tA#vSP
zn6Rh4%{8$5fX)!w^`YRrQ^S`YxW^#$YdFCwJ0&D=%o_)uF}#X8DEvf^fl4#4zmNIG
zVEQ5$I+;YJdx9aFd0|g-IrWvz-aCA3kQ#(0L?Yp^Jm7k*=LFQ6#0vOWQ8A2!TYM3j
z5=2Zi3p$#=0VshzgsqAQnF8Pk1S3tbOUvu3iNK813zh@!&5q`m0AXbE8VGScCvpUP
zu{V_#*vYv%erj&qTX4`0Xd@~VR(&i4!V&Ni9C~>XI10>_q*yRHqa+X-7M6#{8nmAV
z#0F&2WqI7V1wz=_c`{!EacVOanItBm2*iRzLvw)yXOuy=oBI3ks}2dMC$0>j6k-em
zlG_3_zH?{J0N9gU8Ad_!k(A-xr+$7^j~m*ETV7-(v_yaE5bP{f_~#@*qnCz5F8DC1
zfO|x
zP!VI+d&o&{fN%n;Gbz(*8~$RB{BP^0{{!JjbnyQdXBFnU3QKv5^NHy>YbpS62uu0!
zn*J+oPBXVxmtXsnLCVT|2hsXe!q`XU?uvAw%twK+&09cI!I$Uu*
z^-0M?kAehE=%t6CC&tGs%j+9?B>b|5vo($vs+(|7S*#9q*CqizzQh<`HLj^tlTQh5
zY<}8RfgwNLO1#Ah&=1g3Xi(4@zC@~i`|!O0Ge
zNNA!3X7T>hPp4?pvLUj8xgAY@=B*ejg>N_kWLJSZXL$Sktg`;D^#^|}>
z2-nG_;SMM8!UbpTHHxe}^0;^an@x(jwD;~kdc;cIAtaO%N--^PjhpBubJ`FUqX+b%
z0a|^De4GDX+i@(x>#m>nj%fqo7bd|`fUfge1iB_B=EOWJ*SQkt29H_UQv6$Ap9a(M
z>;DXf?kdxMjpN&F|
zmLh9h12E(UnY}wbyN~Y%q*>1idHr6PP}DamO;>vWGD0HDU$ZbX52cHkKS4?p^b|yM
z<~02sz#}#p_Y53@1iUUR#U`o-$ttjb+TGQ~MD3*c8XO+ZPBjw1h%GV}F}`wF2PO|(
zieyg6TOp4?s~jDyy^2qt1cLdCxPP^T#cpTka`)Mg4r6e$A@oo(Joa)BGIVKq;!1fd
zPoV?mm6++=?bFH`#c*6f!H<>=+)VM2`0q<|IK6|1N9|kXisByWo}QlJ$y!r$&K0yr
zjt=JeD2utkZA!kIepwb
z@EQ8w_tQKTgrFe`41S8+`F9xA6*s#9M=RQuVoJOX_G2G1(wKu$&tji)Y8%5$NA&%c
zjNeFV3LJu;x#8tmCFQ_X4CItpG*DS`CInI;alLJC{!~;|4OIO#-=aovy!#|Ai1bcR
z(|quX$5(dn62;-W91ZWYDV2S!A|T$dPz!*MA%MUQCvY8>f)IG7f5k+ai4@yRH$H%O
zpH4$y>vchSnflpZjmny-1x^8%N;4
z!jZ#bE^v9awtE30|YLU!_0v3)vcv*jHA*z>&Px@Rp#*
zKv_@eO?~qxIa3B16j?fwO~Y7={`Yr(t|Ws(Ab3)oK7^?wRJpv35VZ6_)>xLVqa<(`
zuUBTIja0Gq|A!ADuQ+>(lT5Nevqh9&41rW!vrQyM6UjU$MHah@3~oJQXJCsX!BE4j
zeLiGT2`YWfgiY(A=ent<@H5NjW(uR7#Q3gMi(O9Lcz>|}7cS#^E~p*J^`U=My)GsP
zfZ^z5W8~A9k~2-@o=<4Ug)wxlPm{Ch1&)8
z`iDuzYl|9;Qs6*Ydr7sbc403ZtzcwZ%oJ^fdi@Q4$!6IZ%r0KSbaQplkGLH;I~g1|
z56LMcg%6_6!kQ!=ERr8c*1Nz>6UG3ub1C8Qh#bd;UD79Roy{Ay3+hQz_4_x{T=u@=
zlZYS&xMKP1kOjkyj&PQs5*gr;^}W%KTl*Jv50h*TpM+~dvGs+eLi5y&)X|If;szlm
zMUnlv2jP|8|2Z{jB1KR^owjc1yiT8t12L3Qx>y4`=f{$RS6#U#Oqs};eB21ypcma=d?;uf2|of8rXoQ4mdKT8vt1*tId&oPjq
zcpKpj_-r`fPfD!LC8lyi!ZvRooT0H3=NP&{dn8kaFp*)3&ptRE5Z}9DgCz{KCXCS|
z2#kwsLUC-LLc}r*Ly74JrLQQe4btrN@xJm1LpY1@JiKt7XpqRuV8xI##K^tHBRGhW
zbZ4BOauAU-vZ%0*n<0@x{US46;H=O>4JPzL@!>G_wkhDw$?f=|wN!$Ql4
z-bjw?hE_U-97EqV{Y!kfB9eAOcs~{S9h5%&lP1V3;b=D%jE`xQ4ehE*dhud}91DS&
zDD(*HoWa*{d34S^U#opO6{m*4nY6^wS=j7h;*R1SH}{hRW={rs9t}5~PiR|YJt0GK7jC#Yq5sthkF7(Mz+(X%i_1*CY<8(2U{5MH;oGv{niE
zjg^*0DF%_u8*S89f)ZiIXf^_Fld6J3FJx!H7{rU_AO{N>=gwfc^kOTxr$P`T$5xRL
z0^Jil9tbeyWo7!g&ZCg2QEKY&S%W;Fhe$%RLa`)36G9ZUQuH$N@9Y?2Ooof4yrRMj
zLS)a!hnt4eAguwOk=u`UJB#6ob5MVyugxK8LuQ~NBO|dKh=vR0x1U|BSgm|+9Jvq-
z*ThCgr(diaLpYjRRNVZ7(f=6;0DvPpKcxWoCJQ7YkNHM$Rm_aY_{!yi2&16n0AjXo6QQ{av-
z;H@tJp#yb###<5D690ZxHjIwtV|4PCk~Ep*ni4w#W%8kaaCuqagUONfC?L(s$YUqM
zC0!tkqv+@1A@PP=X+C8IM72jBERr$zr3Bz2p+#P!APcUUdjgel&6X3hD2o^jnj--;
z^cP6E1kaEJ4mp_!iIq`&HXfW(Uyxn@x95PmV~){b6b$UflXgu&B-bb^Xn%Reo_vBh
zryERjE$0HVoq{SSZ~}~5xj+UEE*3I)&A&8nA2*U{TPV@k;`FC~T%)~M_D0=7+vw(e
z;x@H&Ww7kY5AI
z`IZ-_k$mB)`i9F
z^Nw{Ye)uE*uDmRglT3?Q0s9^Q-waBx
zDm^!z5*Fa%BHhjfk!2iOq`x5NgmG&aS%pqv5gV!X)yU$YZ{5FWU6b4d4t}m8d=aIU
z9G?25khB;Zua;H3|5s_({tk8ehE=;(^>rwzjgUearw%JhA>*tss~jW84op&Hum;mX
zi5yBP8$zUvu%jl2aYzm&rG!?P}{r7j3@LczXFR{-*JD^5MLvYE7%#agoxn^@{TI7il;%Uf=Q
z7PGu>GB~*Bfdq0tt)z-g4tHFshc0ytY-e!g^bju*FP6@3a;~+EuVjgzq`i4aCm<;E(~*hts_DOxfOnOJoc5`nUxA_jrrr2f3=dmxvI_^%w~
zmSFH8kBi5D^)=(5#gei977sz@1Q(Ftetx#gOM
z%dv{duMiCvjJS-%oEVeTW;@?1Q<+iPj}EXvcen&7(~5JkO2o>=VSRfE{Cy4ZD8Sc8
zsfM)SAi)AE@j{Lm5;6i77Z8(r8GW&qrse=}xh&!%;2wb)Pz-Qa)tq*?Cfn;F+q?;l
z3LiuY0aIPL^B!e87FYmuAyH?-xfm)?Lx`}!RUbnH2?rq5#k)+&s0+^6=e~$@_@YyY
zP>Np3G$IsE)NhZ}ZbZlZS6kNj%;7)}D#d-J{aAb(5^Z^0a|ApB!FZJLKfmQ{he{nZ>S5lSnGoPPtH7H-aAvNe`2kTE<_ykxI>NO54=Y0ri1muOWnbA`ZXmjf9x#;G8?PvwnmPX#~8ru#49NQ<>P4gBK;aoLwRxB{e%
zf#t0@Ya&8#rlaiF1(GM(Q0T^fuIMiUWJ;-tP+_shzVv!yt+A-)FL>GQ@3(&cezadB
z!XiL+4(zR#m|I-Gy2PWwF%SYFf82WTtNDsPa--y%vDT!huNgH>5BhdNg`&aORml#s
z_;%UJxM{%5Vdm$Q=|ov({0Ob)YBT+Ll*44TJ>Pn;+U#0DO&@(>`;s|7aZIP`MzMrF
zvkF)=Kps>n9zA?$(XN^Zb48SzU}QoAu_E^4M*AnfHph8*d!m=EB8Lv`Yw+|~ytR|p
z5&B5#hJAT7C@5-@_vVse+I|&QASyLLnJTU`FlW&f=C4CfMYV@X$i&8#-iHiU@S9Qd
z1bB-*%!}!>tD?$d&~4ti(5YupelRzfc&>id)K$fD8ae1_8TI547M&x
zu69$V>aaB+c8!R5R56@rVu>Ia5)?D{X^Yy$3X#)h(CQZ|eRqTF2VKHc^*siB^dT=?
zZafD2DaFhK>!bfbvz3rf@IoimT%K-U$rffWfAoBwD=5xFX1!Ao0$Je*uuJtPhah#t
z%_0-Bm0GO>Hm|lT4Wx4ej_-cBVukVx;-rPg5?QRf4H#ZPJx_iQv{uxd={S;z{5bdo
zy5caZw}f)9;45Uew11xR8P!)BGOpqU{9
z7y-gC%JCx#JeBy^JuzmNIW-+&ka2@4WW90!rr&v#v3Fa()Lx_(
zs#?Lf5DW@cn{?1Q>+v%v6KbHe5!MqAPXc0>ODUg1>sXHf3+Xj2sq1slPsaJeqUgVk
zmZk_pGUWzxeB=i4v)&PKI&wTx2olT(P=DZx`~><+9v%!5P@HavNTD}C$gNTPEzw9A
zy}W*0#zi|D)O&Ja5^fXV4-On2;vg;rpiyi2dmh5tvP4L2POsj6|WnmyY`X2C+BwsEFrWY92=HIdXi)^}uk<;Igg
zeCqk5Qul{BoV?WVIevT3J@mD5_F^0H;_i8JPPd8Er=4?@#!k{sEm{07FMFY;uZHfX
z?BiZRKc(8-A99cSVVPE{TF>-M;HK($hR(o=;DDzOJTfC&`u^GcS-sDFxM6dqoSc5Z
zK>?mG&Xn!$!VshcIT@4`3J4D?S9*qpmDSYL6cjXjv%|u|qEDYD#z~77En;$SNkIl9
zJ#XGTKuU!wOZo;Y`Ou}7$*WeaLL(=f!C-`kpN@%vYVQuh__=dZ
z@pyN4Hwo1KHC!&&z1T2vA~-a36xXFOmAY-~R)6g@Zj~%@8Kf~UJ359@jFAu*Cno#Y
z8hH84+A(4dYnjKfGnzSDFdKCmw`va(@#jJ(+~Uqoa^ADPZ1c+-P?
z^}|Mb%Ow;I+;;99!hs~)G}&g3UXMa0IAa&0$zY#*+q^luanKn`Cc(j9?K{DsNEave
zgk|PHMEVz+a$9}h6-T3xooeL20Yz8U%T0S
zSQCBtqr=JCteyEtV$;r@%j)b$u@21T=oiA#g(4zr6?D%KriO?}!|HC~|3%`=iKqat
zB}qbtD)}LyNg`+*TZJ?Z^s2&yL=JS?1j6zw-m6C_lEAP1-+$=X#8rp
zNFWG2di0+!U!HvkMvMkWv9PeP>a3wWzSyitXTBD`mpMHo#LV0rUg=3AckYrGOq7V$
zb(1uDbO=18Oxt=DZ~IaB!j+D~lc=cZs|;VxZa8C)etu*4>L}_u0LhrzfYGySl$G_S
zu0+>*o{5cJFPo6F4XO-p_MOtw%q?^lLSJFw6Txk~wVpVh`>_&nNy!I|jgLNy
z7ZGn!I?x65Qks`%R@`|bH}^5J4qxBDVTglJT31&GLkm0z*@VXP%l2`015bPYBCBEv
zLgk*)fQ*Pt3-mPct?6_+gTY#_^C8isIFer`r=vU?CsUnYri|^N9ta1m9
zgRgemSp$w*SRML1Q`jcvaX1cQ=VAN~lrr)wY$L+QeOU?0Ae^b!{{elOW#t--wBhjzBfA
z*XD9cijGg?f0UvtsWGc}s9xw2lawsg?*^MO^a~lvXLA~-B+GMPU;sfu${4oQErYY;
zpr-rlKiPjfeY>)%D$Ab2RW~;^ov);^lniJaXayIO7fa(zKm{vf?1oBZk^3o4xRsO(
zpum-ol+-ViMLb~6!MQW4HPcKE4uJEFi$YgN$3CvV
zg~b_u8!E!{l~ypL_M!=o0cXpMSD=ssJ`1WxpkHt*tq(J4KO3z$5S(m^KNcvFctF}&A-HTu(N};idytK*#u-845U}nQ&Xe@MMU1!
zenfxnLk@p>dIwFu^Js6Qc)0-Cy6dcsn_Jg5Z5J`6h{#CWt6N=MUDF_12>!4WXGp0Y
zQ51?#R3kuJ$;+kSq%k#xp9*QGbwJl&dGS}TUJcxa&>NIfSa>LTf!x91Qi<{MpIxa`
zYUvd@1%+xLx5masJeDh0uAn6TgwLPGb&M6}a5xus;?N%3Kt9+Ke`G;HIoa8@gU4_&
zz8pNm#)l}kb8*@H8z(*o}NA>1)rXf(h*p--W_Q7{UyB0UCe9U&lCzpm1~h*QZT
z{zQ)g*$m#=Z&TXpRjbIjg0Zo%LPz!JzXt~cXtXU+Q(=4d@X0V9zc$sM)bZV8V`GtY
zrqp^mqiBNRS&jub48*n!3X-{V=RV$lyVZAv{`Kl=gVn2j&>@DmAklkoJdkL0{o|)k
zw^*zR+=!>7Xq+uNIvut5@!8XfEw9l_ASeLMMe_3et}ZuMSN)LwC)lk_l&;WPGCl(Z
zaZ*~D&jX1BGcx#SPkGRF&_?gvX%7jMluVkh7KCm(N{5I1{5AuQzVP&h5Rh~~P>k3Y
z>PV%YhGBRUg@$<>vI!-m92PEj#4CFOdn@5JO}
zgXXn3x}b3{ZSPIXi@AD5yn=^z5b>dd96(G7N7h0Ye>`0d>vV59W`soHhPgWp)-yW&*bt+YtMJcVy9i8Z
zV8HwYX9{F|aP6OWcIxWs=^g!RM~2NZ8JQI9d2oI5wWRR#7ZSe>I0TvyXd)uu^=|+M
f`Tc*5R4QlZFrV|4w3_I<$=|aywJ|9&_B`=F&xCNj
literal 33724
zcmeFa2UL~Wwk3>Zm7Nx^C2_E}e!OzX_pmv}+36?1*FT0YA3m$B%PJScS>CO#c4Unr&(kVfXGG
zzf+DkDJ=gGwfq*Fu>N@O$DWu+x1{mORQx&q>cmUWU||EDM^`tC
zoe2yK9Ph8UF1o%R(?X%`v22|0+*EK#RrOt2+3`$^;~C!{KnV|E9+Q%Cu`Ga>LVK{!
zJ6PDdDbtd^xUi3L4VzGm@zN*lJ^yxDX=BEPHEY(Grs`Ei$W?^*)yC^GQuz3uItO8~
zd9H}wxso%|W6_E6fp)x(>is>!4srXu*HLJq`O^;{KYq;q=d1Qw@6zAblx59CIeq%H
zmTZvY;Dt+>I=Bq#d(
z5&iUYA08g?o8|tgCR8U;zjlh5f&G3-?(e3!D++C!>Lg1iUb$`E>eZ{`)Hh?8&K+YP
zTYU?|!K|pRu3oHp_3Bk0uk`eEb1e};Vc|s6(p#(*D$2^fetz*{R(*z@(NdHjO*Y@$
z*BN}Pw2M38(faGfrw$)JOrZq?1dx@F6|9Vx7oU11%*7w=q5C8I^AgUBTe*b~%erV(
z-r|x`WWrQhX|2i`8tV&0y>fO7RG&7Y6B{eSdk55!JG&E$nr-D8(Zsp_4
zvTBj0rJSvbux!f2Zispky2mnfTFiNBsO5#jaMEyxf~wi+(_0s-EHE)H3(4Z_zxul>
zYVH1_q{XsEa1>u-V-?}b7W~eiPI~h$YJOGtKDElGi|RM2s{vHL;q?g9IzW^;1tB
zV0_>*HB=k_GlpjO_M1}e*o!h9v<-@kqIu&Q4Wn3VqQ<$Bv>lG$+WWLu;8pB_B^xfj
z4HdsI@F_K`<+4FT;rcBHxAOAljaABsovl1Fz?1DX_T})MZLDl;m;37yt=sZzla?;e
znR+ePWh}(R-D&Kd1mCTxfrdMgk&%IcHJHHtWmh*DOJaRjmori4zYR1zlW`c5H7gHH
zyO}#bTEQgyO+l7hO-;?ZA+5XQCWoJ&pO}~!pK54eP>^;~>GC+kY(%1K^9>q0z1tpT
zXJ=2yvBktzs~*G>wcMq=Em|54(3JdV^@^%|v(8zq;
zX%*!nZuj*)UP>K%D<&l+b$^egd3{og=d$G@*`EV$+_=H6vor|TSLFS^r8=SgIEHAkZqRs>;**5pWq8YES<3Uy~gE
zSHtoDXxRQY)UxgQy#4kol?2LC5X@Sk{2<2$q53|fa@I8-GL
zQ)+5--6E1P%h+xD|H*lC-H_oj)lBBit+r=oY2N0U1pmYN`=7Y=e>~FP6F-72FK?Ou
zG!McQ9*lG2O?%@3UDJg=j{e+F$tAP35_L`obcNVdIMXc}(%hAPct6c3xh{Bqejazc
zwj#;s#iUO2!=zI6rpcY6?9{B?H*Xk(=G+bru6um=4znUs0HUlLlefV655J8Tu)dU_
zd3K#BtE%!Z?V6BV&He`ohKZ7+9>zna#=G6E&a-dM+HL;`VaLbwQB+h^OiWC^_r4%T
z$zKIz_xSy%zK8>s_D{IKOpnK-b8K5a|CZ33hi2SxS
z1&g!&J55WpG&+hlsLdCPx~tE20VGE_cKYPi<>Ei9B)e`b=aNFeI3~gF{Bq~n%CK0?
zrzKq@K#Zr&%q*LqUz!|f05-cT>tutAch_&@c@x8
z&CRn50esBY^5AMQYt~A)>lL^!EKIfW_?L0&39m-(1GJ35MY}_?
z&VM{pa$_@VMToG$<+;HuK3?8%n{s~Es&zn{-qiV%gGX>hhL{a&otbX^6JT>96!71Q
z@qt>TJc%CfI!G}!@wj6JC(xKmu+SS*}>*F0oc$SYnSU7wbGsjLv6_3=I4-&_&h$`d2yWt*jp@y?ct9F+
z{&c4w(B9+6%MECwjTe%yGGD{{9+$QDEH<2t~^HlH1&5Lqh{Dp7-^kpR|3S);~p+QZimTJucp^UL8@vb2r#
zFwKrWf4aH4<|f9p$fJF*HSYzA3KHXjZN{O
zltAj~vH(=2d(KyX{(MI*X;vFq2G!blXR#j{xjdaW<>l;h&UVG4z>~-^x_MjJ^W2)5
zZqvuA-4<_l2S2%eNkvm4bocoHMoAPEle1K(-3mXfyY72!F7y-N{x#;mK=1)7f#6*4
z*FEy-(VU$c_K0G7w9R>aD$t#RS1pD{#3%>F3HEZ3pl+~C#;IaoW<|jK_*Q*m!)KU?=M5W$?Y*ef(UhudBr-;=k=SC@kuC*U$>mzO?w}$62i0aRS9VRC92&>
zx5);)!K(Rr>htH4Vq&(~I^6=;E^a7^Tv{&MP4v}9y3d}E8J`Jsl+DL3Gph)`%8Uoc
zZd8`&mY)1%YMnb$qJN(ipK+{--R|Xj#b2cD&SAwwI(@kp78Zue6ARYrrRTB`Lz}W?
zmK#I$@-MyJS-N$PNJ~6s%hoUSZItw3BcDMyMG7IN8pSU$$%Un@7J3qj-i}Aw8C0LWQogf8uWf4UciLH5{m{
zihP$qaO0oV!P*edN2w#aYdu#~Sbtz(Sn02%g$w_Oh?{=_kpE`@$Ug?~f4cxKXtXUq
zmkw3{OHn;2I9R7`pXBzp{TxUM`@c-QtGP~%ERoJAkxp*Zn2J=>sHW5DezCE!AT{ip
zUAD9buMnWLOX^jB{0J(jWy;{#lu=zK>SU7YP8!#C1zV-Ojc0GJT6^3yJk%h^W@0)|
z0uUoV=T>^jtO{!k>kqfQu&sr`k`2wNCeMjnFtD<+a%1X<`c#_RST`YQ-t=lQzqxvC
zut5&F%suk0y7mw3Ha)Kt?AeD8A4(6k2ph>+HXU5F^|EGMR{Dy?zqFsX{oIf2{XJ)l
z+w^WuY}u7;{p8^M%*l~Th;x9Ye)V7#;tMuETwa45|iB?9+_}34Q9%R(L&v~?tk<^pFZIh6g%I=9+
zL%G2%=R7moy=Ifh+sexP!OW(BfIhq1?x1(nszE^?Q3P_4n@G
z%h(YYpOn)V7q7^OAl}=YZQJjpcPx47q1KPR+2T)QcAgw?DqO{P(W!HYeaK*jC6r8`bE(6Q_4x7
z(YV~}TebH
ze?FAcuajhWJW1bj?czr@YTiVh!OC@f(sMiBmUz65#eO#CdBw
z#^!nel36~8VG8ZTBPDO50DkTArP)a_R|}097c_>S&5g_|s26VVvhG
zOctx-b(7Ot2J1mjRmlwJf}X9syR!>-{jRDCS#vHMPth0KqV{?%&H~(yMbce6u5K#1#l_)Qj4A~A@GjEf
zwz!GmwgOb(yS*IyVxRh!-<7gG!|FX7Z?e!wCT-X0H=1g41nvMbx7mGE<>%)|HP0<&
z^Vz^UPdn%!2vw2i?Z5X#ddSnhnb0Rp=nHN1JV08BHrFvcM%9})+eJk30X&AEcipO<
z73Y~(m1wc+Ps|zuUk|b!BiDsRl493eeRa(y`SJJrwkX{?k`VG>Sz6h4?48r(nHHzJ
zC^WH3$*z>w>r^&IprG;Qai5-@=*Qn3{Yfv90xAF`WZdVcUkvBVu=(*SSIW-~0JAmQ
zR0tF9_492d_lbCu+Rk&JRh^4jb!D*tqg&Q;%TC|Gb>{}2`T6=LWq%2sYV#-4;)Aq=_()nr?(H58CJPHj&
zNj^5Z61UUo-q^ifD{gKU!>-H%BoiOLuD^sA6i&-7jDK`;a-z_RS|6@T)w8IHQK!%l
zRe2+f8$y6suzo#+Lin38)<9~El&zP)#vFivq51TvaDB`1v*F^G2R<EFDmG-lA^!x#Ga_%f5H*nm}z-oz0&LN
z8{Bai8h_UIb3mXhP?(uFu`u
z7eJJ)U)d3f?i2-h-`=A1Bsw}st`-5{&QWQ8+77qbaUeUljQ7}EdBIthO?zl5fUAw1
zR*aBv+AcUn%DW)}?qdu}mRLo}h2G-hg`pnIP(;oAn~&a$%Yz7-anw3yMMk7570xg$
z7I6#T9KNMUa1E1#@Z(Kia8-^i(joLSKB!f9jgOMLfO4+wy#XiLzUzrY>nC)
z(MgdJuL{aKrr+^?A5)8Ncy->P$)-kkH*8QJFML>C3V$CsYJ~3Jq7WGnp$$460b_$|
zT&Fv{K_3|Nc{Tbg?_pNmDn6&JZ&~1STPO9X@MWcarUXJN{7AA3ip6B*ce*BS6K(Xx
zsaM=;+JX5FnM8vEmgb%B`EXbH^qDi;){z3&t(IRdNRL%x!(d~E3MYv82iH{9rz+kJ4^s68l06ps^CNa2NAM!mB%AH}`I`?BW<
z8)R<1T=&b46EB(Ezj?W73-a@4Lkg0|k|G~zl^uO;X&tewoSB)az0l#q7jbv@d5z*t
zv9j3fK3>>;k4D*~9fn+i^%KCqUx6uaMV30AY`^nQH0Y2!v3^T8Xxs2|Qij8+lJ-bzbB{y@U2P5F2C
zPO`C49$P-hnhVK!na!6ukaeiss4AAq1vR;e8tkZ#suEY3SFc&4bdw$Q?NHiNyi4w6
zA&pd~qomdq^ioT3^(pOo?akEI?*b9k)?u*GUcpjBVku>=H9A$c4|MA1O+njL650@$y*ljLBZKt8fpB{VS$c79xaz)y_5Acm
zPq>3I%HBKMR7n)qF}qpzDb^_Ip^N#rB{2atTl?o)#
zPE=)S0>VeaBrlZ*@)OEneg-i{!m`m0>htoI>k^c=V4dDO8<9PbRzb2ZO27+A*-r)<
zFm-M7UnE#1IAhD&iK+;Srv-TyD`FmL46uuyEztG2^*GUO-bV!u>8S^W6qKAw-JfCf
z;xa*K*NjXM8>^54K>E0$u#bNHnDc8lau+sxWo0FaK1~@HLV|-YgB3BfeJ3ucbGLRM
zHsg0MK*DqEviR1cKw!4Gbp-gEz&%)U=g*(-?d=8EKz|(6*jT7*O8l-Q*6;sxV+@`M*t
zps9=!i~*&V>`tNMX`ho`48^rfMgmn1Hf3d**9wS=4%h0Mw4|KeS9*)*=Uu{1Ud(x*jNOVA^pPs5k0tA52=5!fIEtq*{4GK+J@CFZ6eu>ky1^h4*HH1q6
zW$2W8(7>J|Jz*S*7CJ2!4zY^L3d=^EJf
zpdeyCK^ZgOtL7oV&o32K!oj#?g}F@(6T;m79z8N`y>*FAW+6sIgCZM`WG>wWg!;9d
zlHXvdI1kvTuKuC$>Y4)j!f+M*9f)ZQvlI8_-CzQ7HaG8E2{y@#XTBZLR@7w>2(}gc
zGIHcZ-c+m1{I}Qs(tR;-M3B$}IDpV%`LJ>rLyVl{-d!F96sy^L6+tN%l?t?LYY^C^
zDhcQvXw2wuwyollcb)bV&HqL5D*^^S54S>I!z4c)buh9fALuT|q`)2V#SkC^Oc}!5
z#IwkS0YU`t?Hh_(+9Aw`;cq#|NJ+iCMNqK+tytC1bH+vM6{j&W&I=Ru2y^wI{{bjN
zjGfQpdEZI2$=ZB=gBx+fg7Fz$wP72L*?pnkq^GAR#MlMM7dbXg$)7tOABRN%;5zj|
ziD#ns?X?yU6D`QK&
zl$KWNcrLhq>?dTp%~Cd}A8x&E_cl!GnZ*uu9RQd`z}ejw>V#;21P4zd_)GEfnqmA=
zYhn3=HeE%5r9m}Jas=?~T}6e#^NS}t3sG6{MqG6oNG;8=?+@NSP#s;7aQs=oojbt)
zvk2(CtNo?E+$GflZ3h7LT8=$$@9x|pBl7|)$X{+!4``D2<8hcSeoMXw5GK_Vw!yEM
zr5k7`7iH)zER9fCogmEYR3N`53Q890Egq`-`>4GjN9+!+|Bb|%U!7S5#jma7($C51
zK)|uCjjIJ^N$AczW1Dy4^??gJ1qI`gR+j40rvSpiPm!ol#ZqXfJSEiwrt25^pN0i&
z7qv$LLy}}PD~tR1BOW8ij;k1IRH$<(7dRE4qqtQuz3t{S+xv7MW81=zgPQC}(L!Y&
zn~}?4Mz=D{a=m9~RE^d>(pe%4uHQv6tqc_xq+RR|(hX(<1~;-PgNuliF3GyuhAVF)
ziXEQ8o@b$=!t$ojpwwtvjD0foUrT|lDNq=YX{C&P-#AJpD4G?Ui7n}MX4ViZm4g#0e&oH;g_74{30fEnWa;P~QsJB4SI0$ev
zX)8Z}Y{00aY{c}=5tcg;Omr>8)~^h7t2oT~0HGQs!eLHQ<0O>d$h2%?Q#^;2t1VFf
zaG_F8aVZPzKhil9n46o609I1|i>%{C#s`ldZ%vInV0qX=eFOkgM>gY9S4s8iXm01P
z4@m8Ld!r#4s80VpX}}sqMl%!XiYwnYATQ|UGJq{bM)<-81<%#SX)|^ZkS{!W_tRd}
zVqaEPHQv=IijDQtoG1D?taEoRl`)G8bIDN7bNZ8-tT*}em>_0q8JI4v`17h*yI3kP
ztZ_`80b7nr5L+U8SRpo>#25;V>>UImL4$k*=%W4;1kLO}eDY*Rbck^d)(jN3Tbxp(
zsioW%Z{AGLp=yBEq}y5yRvS$Cm#RoQ;9YH9O2PT1A=%C7&|+V|UWxP4mwRV*H2Apv
z`M9Z@#m%*PUuaeIVRv9#w|t8%)iW;!WpETgn7I(qH4{&?zXOFfDPy1o0WP|HegD9x
z(1bo)kNOI1M-%RnLlJ6IKX!s5Bcje62rhx%hww$AjPxz-+GKcfZ-YY`_P;$mcG1JO
z4_KBQjq8E|?adZT5xUtG^|05@mkqm5y+{iptfTwFEH-MwnTuMhPRUr2bc?^IsOXgj7qhHoU)GJFcB~6C2BN|5*rJ?Zq}(QacNI%x1FgV&_hwL+3{Plbh6Uv#
z)9Qs)?gRu7YOg}%g+=n62v-HkVXh^A8
zC8(vS-2Q}YKsZ!6dK7Z;7%%}uL|8Xvokl+kWQt+zd5}iz>;8P___jwNz_#=A_tq!3
zf!9G^*vE)eG6Plrc4#O#jb|vk4>_PBV5kX()(m5;$nmtTv}3dK9W(@;$>BEYfQ#7O
zLv~9qd3bH^dij6nvaCnRV{2m*_o)U@i==+&ln_oyt@!2-3b2TbV+(*9gM9)~K&SfC
zVOqsFbsx3=`kq4jh6D?^yp?y*V|_Wwm|)3IAXcue<-p!bIB60Xo#oTU$zbJY2CDb#wawQq4a+EOvi-ykEI5zoM}YpZ@PD1@`njzCF%sD36~$
zE$teS9?DyPE}=l{ioW{XvQi8JK;t4%Bw!#GhOz5NF&^i#cq6!PHO2>r@8sP`vFknn
z9=dy3DGI{llcig3NuEN{1aOFQEdm$|QLP4ul>KrRpcwSer=8k}n90@;V>S
zX~4B3x$lySgx(kJwL?AE5>k)86E@0ro^PF+tv?FYd{N^k^WrNDCHe?sAn-Cl)gYa5
z?L4(V`x*FO?6?OQAKET}<9oIM(=L+a^
zp0mobzXUkiJPzZMxlCim0u&qyQD^;Ii96Jx9tA5AJWn|R+d+Vh+nLKAXR%jhe@#V?
z&dSN1$#wQ}w;C^}6o_Wod36UTt(dai_XXdw?8M=={Vi465ez$=rx{*vg^jop4U~xN
za((%lCF?V@q92>E2uc8(r#7~>!{G6YMt<64xC4B4LPCNEIAfACz^n2gOWwE2>Ah=u
zE7sWrH~cfPfOi8?s<5$e{LX#()5+S_QL#%W50}wprS%7gz|d`|559v67eO_z!|k5AH2k=s1g_BzRK6
zBs0QCpNf8ccc&>y)%!b;oWOd>U_*n`La-M~oqgn+W7ivivQkg)xlMODiJMD|^^!o(
zbnaU;BougH6&XR7#e0!iGLHeHQtYr8i;9%fuhw+<80*T+A5I|wQ)q}N+OiWsh*z2M
z3FXnhoDH|HE8~mgRMf+WVkv+$fDjUxX=wh}I{qlzWX8@3Qi@>*Gb=Xy`sfSU+nif5
zqx=))mCtv(u4KE2rN4gyRU`c@TG$Rfzo^G=v*SF=QK+Km@Mtky4&WX?t_V
zlYnB#M~Rmm@YD*gFP)$?V+SfSvYG^e+iQe*z?*3QdLwcBe+>u;aftHj2NduJ#DpW|
z#O!!4P=^JQ01&>Op&@+mQlPiRI6Fkj7Zeaa3MA}woI#*S?pa|@H=YK
zy^lY&a#P}$?qTI&Wa^I+YDuNb24pWKfoK@QJ_WQS5RJer-F#%hqF2UQJ%@7N#x`!=
z`*-evBMkxQhVUc0k)Me4FH3QIZRwT4S0jK#GPG~7gasPiQrFTNM~y~fZp0AbmV9-2
zFSmUKJh*v~D3>WrSOZ&i_RRyXePEty_R0=C6
z_%39*b66FOC&4SA(lhRTp^hrv5ln_|?q}3}s<}23E^TwhHv6(VNre(8I^N$;MzO=Vq!rj8kp)?tKL|Z-XKP$Wja!-o_Aa6HDFoser#{=yC)57Vhm&Zp!#;6c#w$uIy4TZSUgro~96JLO6
z4t+KJGBR^yga?nDX~XB`gEtDpW=9Du{mD(c6Fk8n=m8xaop7U1@AU|!`EqY6pfh-D
zV&uuFH*SKYjFk$dau?7HY*f4|!Nke6?WT!yZ!}>yV6iW~u^FcRVM0A%vC
zJ@9xl#@IaFwasnn9qLAY{_`*(0dEPq^!>4KofOFaDk^>cr!{QE;P=G#SLr{mdk*WS
z!R}EFZ-^m_B80#sV}{$2kskF%4p*50Fmh~q4<7eg$;6F9B)ia2Rvw(&vr6fB#4OAj
z5=}0+X`KW;^`yoJ-rVed#rXHgtCc)I^T^AhbS9wEd3^d~S+T76o*)M2q*yW%d*VEYrl8gn6_GSk;WtPjXuBxX@)h{0Il{a}L!9EAF_##20;
z2dGiyJaTtO2k@f>5(n-PSmZeg5t+Zvlu{RW;xqJwbdpLzbu0{=rJ{vqwk|6_1DOvB
ztUE5`0i_sA!K3l~Q)2prERPs6he%7v(Ty87YAxQaLLLEEw}`_38Z)E~^t|L@I&c9J
z&cN?;?l_^)%~S9GwC($X?l?H0Sc_zrYTBlUHiA!Ov6}q%ZwbmZc32xAm%=R
zcjzO;hmNGujEW01kFK#(H*l{&UBPT$Jo$sog<)s~H&3Af6($+yNvoek>4lowXHd#x
z?0(y6dMT*?@R*ALxuS;8FCzRAzBhnZg?tby!^^|1ZpflOOb@~$7ssF0LomRfa7Z#p
zgT2uqY^?bx72l@3qOTA&q^)#sntz
zn0nEsx^Ww}U!QIP5mb;%+~ENTdGQ!I67(9s(V1N+ydZlPca4Dei}ejRMkMfBu~tCl
z9pH~V#A|RlXly&RfJt5GVDdPJXJw=y6vNDEtE^m(o}x9YSEoIDwqY*l@7c4C{2XHx
z;}3~LB@zog+VmVC1*XEGMqO#p;hT7*JJ;BC5-IJsg1ANmQ&*?N=yvQb^n&>37N?RVP`yKhd~!^>5}Gb>?ndN&9-WwH{X*
z3mrJ%tDyes;m^yKZ4Pu=edQ0@)}Do~+;@D|KkTI4Ia%5>BY6Dma7Mj+kHeGU2A4sA
zzfRE8vCSPE9c^L`=*)Vg<$~yi*wp3G1C|sP%g>2`mTM?P4MlP}AIsWBd6U)VZVC-**`XFL73M;uIn6jZaI9dqMhe=T`x8lTVXk<@
zDRQ~RcOL#*glaV=f4vPJ@Ytz~m=-+%u-
zEq=h#FX)#tq_TaCK%U8Oj!m7TIuy*_^Fjsd8HbpggxF7lq5MF~YqE%!uai9C;mkn@
zypv#j&BSim6cr@3xzFM0$Yxo*5tjgQ6KHhd@tvd2h*~SwvezJlpl(hHF_mBu;ij-g
zJ{a5gYdcnJ)av4oHxVvKA%PvIKHu!SaH;w*U$G3Ki3Rd?z#Hvj1ZZijN({-!vbF5g
z7
z;K|ukFkKubW$Wwf+bIFpwF{A9fu>`%5vtv4wRK^QNBjo)6cl#o6c|W2U+6@^wSF;W
zJyIrFVNBz7-`Tjy#Pt$sj15Lb*zww16rzOjwG|@Cb+zdAkp<5{{6glrSe8NhEs*et0*f|PltwccQcgevq-n5j
z%!<)W3BO=pfQY>WxPt3G0GvP>hVc+392UVzNRr{mxUdb^K=uQ+afh1&kk1|LXs5c0
zwhKW$7N5RB8=`;aqaw4Z!rZn8?qLw3AOS$31m*2gsgWZ6f1U#Bu7O2}CKAJkfqbK7
zl68|?35yGAg)CNNLksjEpcm3%s2PnV^4u}hIrI)0VPbnQ>
zX0k7K6mC-uH3UTi_>zmXCnyns2b2TYf)&OD3zeANaBFbz2+;lr2pA>Z#K8tuFz}6m
zaVkI}{&-Zz5@j5aTKu^2@Ev4P;)Cld^`}C_M+R7gi63=56VyL2$$nMu)l0Y5Huk^R
z-vS!s!sq;Vj?=AuZRbvzwT(8V-Av4vd|_dlZZR-t)lg#4W_5gF%i#p?O4VZ=^qRWwTMiF$ltC$CX{UXWcawCU@ScM1R^R!jBk(9
z*-%I-klQ_s#8=mtl#;SME=e5n&C`K4*WlS93txn~
z|8cnw!%~u~?ThkSXO6hccoL~~?qd}oPgmQ#5*(}$?7t1?K1x6uMFSQZkmy9sfQ8i{
z!#wEr?VaeK=GS~uv`H`Db%tcLNo-nnWg@G5pQj1pJgcZ6j@=KhJd0qIf0T3a-SbhoGl~+|GBKl`OkdX=>Sc@ik9ah5EOjORuwOdiIK)5o6
zQbNK_NC<%ab4=e7U5<(lMQIa87>=(k`JI4fT|I^500Mwde~HVz)n=VVK8XFOcEJ{D
z^VgGNHORtuBd8L^ATybe}3
zERtw!gPKC(EJNWL5)nbL6Bp#XkrF&55Z{8YVA7k)N)dZm6~+!+Wn_bz?0Txg7TQG^
z3b|0?rRpV2C&jpbA08cOqcGGpL&qiZDlo_C0ISX-_TFy>(@|9MHHC{8_;F#}PAO;Hjac!=}bX?f5+qbGl(eM2DM
z#L5~D)CC`eko`yK&+LlZWw-`V5$eWuLleL@;A@Fl|L3%$?#_c>94oK;%a_UsxjEW~
zqiA>?eR*X$KG!gN9UfW=_H^b#f_<9rf=>rW8d(lL_~`qb040@l&s0lYymb=
zW6mZ2OoM!v7;(S`TexC8)}bqn=o3)n0j(F`+}y$zNc1Pyf6
z4xuLx3Kr=zD}Q%ngjD}ndDv4j0H`XNq}706RbofShC&MxMga+AS#?r6q!y9}rIEkPEBIBmf3&a;<^7bmEn;_hCK
z2SR>P;jSlVfh?V#;=1@@zYpIJh`^#QAF*^4@4ab<#_NHN+hsDH3@T1zL`jO=za_n3
z$uOFPRIM=x$C*$;ydy(uscMMj4N+GZOG_tD0S>b^UuECA6eM#X1ugky1yhG$OUGva
zZppa63z3IS@l)4TMFC-XuM%PaZQW=sJk++%VeMS#VPHPo9R6
z1Ze>AkUxJ#`J$gm{bk}fcz6YF)>6@x2ClsyLBF93(+-#b+)nQ202N~ZY*^Y*=!~E)
zKU{M{;4AVk6D1cw8U2Dn4f0s9Y0FA6S~vh@9K?bLye~dn2j}00qOjo2Y?52DbR6SF
zOT{HDPJ+3RR&HjYKqdxts8jRN>Q%fH$
zhl<5g@c#V#?qHfEvK$dGmOL12k-1{D6!8C0A8_8!x9iA?T{px!ZxSm3$VHSbi$HK!
ziCudZIs}b5DQ~ioRZwTExh(4?yoOk&v1m6LJDwaV6UX3W!?D&?pT-{xNTX|Hug!RJ
z`dbH#Z#rd8yHT)T2H8Ytwjhp!t|HZGn@+~A{=T=1;In8$A&Ow0jBIriwnD%KqyQ9_
zZ%Nk-$|6GOprBO|DL{Bfv#dK-SogRTI6dSs5VHVZdBS6$YT8sjF|0Y$b{sGl<`y*=Nx=jp}SiekYVNa@Y%sQ_@Lo
z^YrsT)m@|o8GIaqBTz~^3B>5^B}TGoqY!XXHXU$}bRwZPLT*7YWCA?_$tcZ(zKC}b
zS0M`R&5`DhSx_HyD?SHuT_-pG9<;50NTDA?Ryc0o`PAG=x-iq&G!X
zwI$p3EOPBFxXXZECB(#@O~-h1m14z3J@Vg~Q3l!`zhir#~NcADS4?=
z-e}E}$Nz4bxoj?z@36SzPa!#$h^XQ=ZD(X%ttYH^r>0rYST>aCn1mT+?$qBi>6T^D
zm1v!nuy}iG-M${3LA!G0oH~u_pkuoIQm7P<91gLz<~6d1aSn(~kr7iSxP5!fmqfw8
zst^lYcV?_u`K>5gs@Qugvpv~Hq}3Psh>2rQ!jsZ~L$^pk@Z|ADIDubK3`YauIIyJN
z)U9_(S;5d3h5wL6xeSM7u`nF+q~RDkIvSfkAwsfq8feGpek8ACV6jFGK>`9g1E63M
zi3IRnL_d8y8#$_A47)zL_XF?uCe6MmjC9N8CQH(Ngt7wWezZvt_bA4#<>KqLfW=9F
z^=fP>c{nV@<$U{0*Aa2xpF7H}q6lX#)DY%E%`jBcYx$w0ll_tz;pEYxGUJh+u?{Xc
zZ#)*issq>bd}I5gk~T4eD=Wt>4zg*}w{1wovJsb~37jktP&!<%KN-(bV{56S>!5l#
zBzcx+3b19G2lslb_*8A~gmhGry!I8^#UF{Ih1uy=dyD}1bn$@lvs$>6Fyxje-p6<+
z(@ccpT{wMoYW+h3k}?AAcw>4Rj~v$hJs^pOC<-9LTnKDV&hT>TSI)7~&%80eu!73G
zG*y4;s{Uu1UH?HN=3kuU|5XR${dJlBug2{Ee~xOhu_{yh9H~GNZr`9ND=R~p+zmJ?
z3NarrU-xtie27HnL`WmXc&uM!%=ZYc=%^V=|9Dgd9*&G}fDPlwSc9pN{YY~7aZ{Td
zA&%k>S~5w3M4yNDgDw=r_6(~QXD|=QsC6V_!Izu?L2KTgl()nmy68aUO8rKB(4U|V
zrtlNsSX_=A#_4<^@Up}aYul;@;#Z3jsOj@G1^t@&K(4>Yp4_yjVQB)msVG6N|!H8pVi!h__5f}K<^Xy~*@^%-c?kG6#8$n)`M
zyh#8NP>=6vcLWcLR>&1U|8fytmC~y1=OKVY8}o>M@&q*?*o>_C+NSjLV~~A}a53ua
z2mj8MW?(_iL1iYbsEj8;6(DldrfkAh6}02?mMOWrFCxLAx&Urb&g8lHav3<$pDDyK
z-aYll?b{#V%}yJH5R4-zSy>?{$e(;o_%!pQgK`JVkPtx_lJeepLcm@VG(8Mu5zyDN
zBhbh`?+?;HFb|!lv-K#D!2W*$qezrh7-D&d3b}_I#G&Qi$Z-)LvyJv1e}g^IkjSuZ
zlc%NNm0-+2e|D2+G(+uAp&?wC%<-Fn5gVZrLkVe;11N!&Cac*-rQx1Z7mn>iDXNz#
z2Ucj{s(#G%9Or_DWS)c^=S}$bMB0^JQ)?6-KNqyU)`Jdtjnz75T315g=;7tvF&GWk
z-w`Qqq)8MfyKCs_-MxRm2}s1P?=qGQNMJ^aA^e@i);lhN*9FO5MC>@wM6n{<;U91T
zI0J-SfH+4b%4GA!*W5CM7vkpQU4W#DUV7;&7o{614|(yODJ(+_isq
zPI}kZm&D&YTHD^0U#gbvnw&GM^%Kj($rYTyk?SZm@mb#M*?5lJ4hhLpTULtshO^)q
zm@L?tiPRiz#RVa~%!5h#YZOuKK+6-yFFx2CgWi5+-k}
zN?laa*ViwWdHe1i3^y+Dl(ay^C7oW;>)JjMf?ycH3j|gLut|QY=N8y}{O{O-fpI+e
zIzKLcs>ygV-re4R%ZI+ct{?^b2@#EI>Jj_tMRk0`ID^eCs!9<7jPTLa3bZpTD!9v#
zVKTT@va{#Gz=(1Z?CFdYyeKd{!jyEaV3j^YDTXHmp_9{D>ITCD58B8{KWG!LE^4Gn
zZ0KN1tSDR9Gk(enP_S=5QOcg(7WWEwVn*&HZgxbYIwLqB>BkAf`~1Di$Et&{n%C2P
ztbn&&=6RESL3y8R*fl!EF|Sb5T2XYFI6G57T7dP@kSOlqu>@BB>Fh7)3NOH*5Y&yJ
z&j1~r
zRPRGIO(g~pMMf;3#;2@^QJ&_SF3WErIb#;RnV~9ZrMUs{;8B-Op_z7(0r|!Qvk<7G
z29UxaC4$BJM_Vhe%tQ5uB*m=w6$F#~&86e0_GC!3W0k_phLa`8qM)tXfOg4!j0Tw&
zPN#>TLgj0G;;1)fB0J`CeyXE10PvThM$FThKUHb#T=bLy3uq!&~D(ZP41>0nhH
z3C5ZrP1GYZ;-tv}_p;5Mi9Esnn7lhT(2{D`OIjz0C=V6G1LY`L@W5jbWw3Eu(eEA2
z1`e<
ziiJ>a#h?@usvbgbn#oe=jg5byqdO(29!Fn<*e}p(C&U*2C9!^)GW&LK&qs0S+De#S
zn;_l2RE!`o|9i#A3Z5xs#Lz~&$F;Rj1^!_E;Z|H446f`AR^~S?#2>(?a>1EV`;_P$
z>kN~G_JSk-lGsR+848f?y~^;Mp!#xOI@K6@U1(O)V?ibMpVTQRVgMb)Z9bjknvS_0
zqPxj?L_^1eyK~K?<8G<6zW>f8r^>^g(_!_
z26SLTxTMlieIRp?>H~)rq<{m6`}%LMF3{(?2-C%A?9&LGgN=4i4|v_k@iqi=frCRb
zC3OT#>+3~p!c@ep&yfDRngA_|G-jOTJwi@H&Sz*(@UNyPV)rI8RdBqoMhWN3fjk}tU@
zJ#bJzgMw;o-m=mq`#wl_6?n_+HGvcMF#?hb3kWD{{oBVrA@FL6YzL`S-ny~i+=Y`T
zPvU$k48?}(ebEwGaMVIk%Nx+gV~d`(2z&(C@jW29r}Xt1oi1>{+xKljQ`aFeX-M~>
z7JWf8xs^Uww|o)j;(hl5Ot(bsT~3YA{ykttxjFz9AdNT6@G@T1MkREdA}PqX|I<@$
zfxK*DcY(ml|Dp~AC2Gl(tW@prB}88PWC}!4E5o5;`)M|)!SRxds4WhZgW#FcM4EiEbX`%mpQE%U4prcZzHsS5t
zKjAo3Ri>4by5O9;h6`XvDyB*)*hQn?$3wTGtc+hsD51CG7{x{QCRfZ=0#%p1WI$a2
zNh-W+Mpo@52QGvewr-7`tHEs4qX^SHDiv)p2cB|lwD0BZ`&32{%3%6eQykyKEJuah
zxlBo2{dT~Ub=M2zy_rEN1@5+27|E_eM0b(Bd{-YFth%~O@x(ejX2nm
zN0FR9QZ|Y0^fp2c&A$*x{`Pf>jKuSVV=c(<4?w%aw3P4P;9A1-1+0NKc